Objective: The purpose of this study was to determine the diagnostic accuracy of sonography versus magnetic resonance imaging (MRI) and positron emission tomography (PET) in the characterization of adnexal masses.
Methods: One hundred and one patients with asymptomatic adnexal masses, which were scheduled for laparoscopy, underwent preoperative transvaginal ultrasound, MRI, and 2-[(18)F]fluoro-2-deoxy-d-glucose PET. Two different sonomorphological scoring systems were used to distinguish malignant from benign lesions. In addition, transvaginal Doppler flow velocimetry was performed and the resistance index (RI) of ovarian blood vessels was calculated. RI values below 0.45 were considered to indicate malignancy. MRI was evaluated on the basis of signal intensity and morphologic features such as wall thickness, septations, fluid or solid components, and vascularity. PET imaging was used to determine 2-[(18)F]fluoro-2-deoxy-D-glucose uptake. Malignancy was suspected if radiotracer uptake equaled or exceeded that of the liver. Based on histologic findings, sensitivity, specificity, positive and negative predictive values, and accuracy were first calculated independently for each imaging technique. Finally, a second session resulted in a consensus diagnosis being made based on the findings of all three modalities.
Results: Sonographic evaluation of adnexal masses resulted in correct classification of 11 of 12 ovarian malignancies (sensitivity 92%) but with a specificity of only 60%. With MRI and PET, specificities improved to 84 and 80% respectively, but sensitivities decreased. When all imaging modalities were combined, sensitivity and specificity were 92 and 85%, respectively, and accuracy was 86%.
Conclusion: Combination of ultrasound with MRI and PET may improve accuracy in differentiation of benign from malignant ovarian lesions. However, negative MRI or PET results do not rule out early-stage ovarian cancer or borderline malignancies.
Copyright 2000 Academic Press.