Three-dimensional ultrasonography and power Doppler in ovarian cancer screening of asymptomatic peri- and postmenopausal women.

AIM To determine whether introducing three-dimensional (3D) ultrasonography with power Doppler facilities as a secondary screening test, preceded by annual transvaginal grayscale ultrasonography (TVUS) (followed by transvaginal color Doppler (TVCD) in selected cases) as a primary screening test for ovarian cancer improves the accuracy of ovarian cancer screening studies. METHODS Annual TVUS was performed on 3,201 peri- and postmenopausal asymptomatic women aged > or =50 years from March 1, 2001 to June 30, 2003. Cystic ovarian lesions in perimenopausal women were routinely reevaluated by TVUS and TVCD at 4-6 week intervals to avoid unnecessary surgical intervention for physiological cysts. Any multiloculated, complex or solid ovarian mass, as well as persistently cystic mass >5 cm in diameter, in which the echo architecture and/or blood flow pattern was not highly suggestive of a benign histology, was categorized malignant. In these cases, TVUS and TVCD findings were obtained in no more than 2 weeks and supplemented by secondary screening, including 3D ultrasonography, and 3D power Doppler, in combination with serum CA 125 determination. After detailed ultrasonographic examination, surgical removal of the tumor and pathohistological classification were completed. RESULTS Twenty-five patients (0.8%) with persisting ultrasonographic abnormalities after primary and secondary screening underwent surgery to remove the ovarian tumor. Five epithelial ovarian cancers were detected: 3 stage IA, 1 stage IB, and 1 stage IC. Three stage I patients had a palpable abnormality on clinical examination. Furthermore, in three patients with stage I disease, CA 125 serum value was elevated (> or =35 U/mL). Three-dimensional ultrasonography and power Doppler, as well as TVUS findings were indicative of malignancy in all 5 patients with stage I ovarian cancer, whereas TVCD finding was false-negative in 2 patients with stage I disease. Screening test (primary+secondary screening) had the sensitivity of 100%, specificity of 99.4%, positive predictive value of 20%, and negative predictive value of 100%. CONCLUSION Application of 3D ultrasonography and power Doppler imaging in patients with "positive" standard ultrasound tests (annual TVUS, followed by TVCD in selected cases) represents a novel approach for the early and accurate detection of ovarian cancer through screening.