Gray-scale ultrasonography is a reliable and reproducible means of distinguishing between benign and malignant ovarian tumors. New blood vessel formation is a key factor in tumor growth and spread, but conventional Doppler ultrasonography has not appreciably improved the assessment of tumor vascularization. The advent of three-dimensional (3D) power Doppler ultrasound raises the possibility of objectively studying vascularization in a whole organ or tumor.
The value of this modality was studied in 106 women scheduled to have surgery for an ovarian mass, which was assessed by transvaginal gray-scale ultrasound as well as two-dimensional (2D) and 3D power Doppler ultrasound. The examiner rated the color content of the tumor scan using a visual analog scale. The whole tumor and a sample taken from the most vascular area of the tumor were analyzed to determine the vascularization index (VI), flow index (FI), and vascularization flow index. An attempt was made to include the entire tumor in the 3D power Doppler studies. The series included 79 women with benign tumors, 6 with borderline lesions, and 21 with invasive tumors.
A logistic regression model incorporating only gray-scale variables (size of the largest solid component, wall irregularity, lesion size) had an area under the receiver-operating characteristics (ROC) curve of 0.98, was 100% sensitive, had a false-positive rate of 10%, and had a positive likelihood ratio of 10 when using the mathematically best cutoff value for risk of malignancy. The 3D FI in the most vascular part of the tumor—the 3D parameter that performed best diagnostically—was superior to the color content of the tumor scan, with an area under the ROC curve of 0.92, sensitivity of 93%, and a false-positive rate of 16% using the mathematically best cutoff value. Adding the color content of the tumor scan or FI in the vascular tumor sample to the logistic regression model improved its diagnostic performance only marginally. There were no systematic differences between the first and second measurements of one observer or between measurements made by two observers.
An experienced ultrasound examiner is able to correctly classify most adnexal tumors as benign or malignant using only gray-scale ultrasonography. Approximately 10% of tumors, however, are very problematic even for expert sonographers; it is for these cases that newer diagnostic methods, possibly including 3D power Doppler ultrasound, may prove to be helpful.