Objective: To assess the accuracy of pelvic sonography in distinguishing benign from malignant lesions in postmenopausal women, using morphologic criteria and Doppler flow characteristics.
Methods: All postmenopausal patients scanned from March 1992 to April 1993 with sonographically identified and pathologically confirmed adnexal masses formed the study group. The adnexal lesions were morphologically categorized prospectively as benign or malignant, and pulsed Doppler flow studies were measured using the lowest resistance index obtained from each mass. The sensitivity and specificity were determined for morphologic and Doppler flow assessments, as well as for a combination of these methods, for predicting the presence of malignancy.
Results: Thirty-three postmenopausal patients formed the study group; 12 lesions were malignant and 21 were benign pathologically. Using morphologic criteria alone, the sensitivity in detecting malignancy was 91% and specificity was 52%. Using pulsed Doppler alone with a resistance index limit of 0.6, the sensitivity in predicting malignancy was 66%, with a specificity of 81%. If a resistance index limit of 0.8 were used, the sensitivity and specificity would be the same as those for morphology alone. Combining morphology and resistance index, a single malignancy would still have been missed (sensitivity 91%).
Conclusion: In our experience, Doppler flow studies did not add substantially to the prediction of malignancy using morphologic assessment alone.