Sonography of ovarian masses: poor sensitivity of resistive index for identifying malignant lesions

OBJECTIVE: Current research suggests that the resistive index of adnexal masses is a sensitive measure for distinguishing between benign and malignant ovarian masses. We devised a study to determine how morphologic findings on sonograms compare with the resistive index of benign and malignant lesions. SUBJECTS AND METHODS: Pelvic sonograms were obtained in 34 women with 36 adnexal masses that were likely neoplastic or non-self-limiting (such as an endometrioma). Transabdominal and transvaginal sonograms were obtained, and the resistive index was calculated. A threshold resistive index of 0.4 was used to differentiate benign from malignant lesions. Masses were characterized prospectively as probably benign or possibly malignant on the basis of their sonographic appearance. Pathologic proof was obtained for 35 masses, and one mass was diagnosed on the basis of CT findings. RESULTS: Seventeen lesions were deemed probably benign, and pathologic examination showed that all 17 were benign. The resistive index was greater than 0.4 in 14 of these 17 cases. Nineteen lesions were characterized as possibly malignant. On pathologic examination, six were benign neoplasms, five were nonneoplastic masses, and eight were malignant neoplasms. The resistive index was greater than 0.4 in 10 of the 11 benign lesions. It was less than 0.4 in only two of the eight lesions that were classified as malignant on the basis of both morphologic and pathologic findings. CONCLUSION: Sonography is sensitive but not specific for distinguishing between benign and malignant ovarian neoplasms. Although use of the resistive index might improve specificity in the assessment of possibly malignant lesions, reliance on this parameter is potentially misleading, as six of eight malignant lesions in this series were miscategorized on the basis of their resistive index.