Comparison of conventional color Doppler imaging and power doppler imaging for the diagnosis of ovarian cancer: results of a European study
Guerriero S, Alcazar JL, Ajossa S, Lai MP, Errasti T, Mallarini G, Melis GB.
The purpose of this prospective study was to compare the diagnostic accuracy of conventional color Doppler (CCD) imaging and power Doppler (PD) imaging in the diagnosis of ovarian cancer.
Six hundred fifty-six consecutive women with adnexal masses scheduled for surgery in two European university departments of obstetrics and gynecology underwent preoperative transvaginal ultrasound. The scanning procedure was the same in the two institutions: after B-mode sonography, a CCD/PD imaging scan was performed before pulsed Doppler evaluation. Using both modalities of color Doppler, malignancy was suspected when arterial flow was visualized in an echogenic portion defined as malignant by B-mode.
To avoid the risk of bias due to the absence of blindness of the examiner after the first Doppler evaluation, at one institution 328 consecutive women with an adnexal mass were evaluated using only CCD imaging, whereas at the second institution the ultrasonographic evaluation of the same number of masses was performed using PD imaging, and the results were compared prospectively.
The age, the rate of postmenopausal women, and the rate of ovarian cancer were similar in the two institutions. The false-positive rate of B-mode imaging was similar in the two institutions (17 versus 18%), while the false-positive rates of CCD and PD imaging were 4.6 and 7.4%, respectively. Although the overall diagnostic accuracy of two techniques seems comparable, with a similar value of K (0.81 versus 0.84), a significantly lower sensitivity in differentiation of benign from malignant ovarian lesions was found using CCD (87 versus 100%).
At least one of the two Doppler techniques should be used in conjunction with B-mode imaging in order to decrease the false-positive rate of B-mode used alone but CCD imaging showed a higher false-negative rate.
CITATION Gynecol Oncol. 2001 Nov;83(2):299-304.