Comparison of transvaginal color Doppler imaging and color Doppler energy for assessment of intraovarian blood flow

Abstract

Objective: To investigate any systematic differences in the analysis of blood flow velocity waveforms derived by color Doppler imaging and color Doppler energy examination of corpora lutea and adnexal tumors, to test whether the accuracy for diagnosing ovarian malignancy differs between end points derived by color Doppler imaging and color Doppler energy, and to compare the reproducibility of flow velocity waveform analysis obtained by both methods.

Methods: Fifty-six asymptomatic women with presumed corpora lutea and 67 women with known adnexal masses were included in the study. They all were examined using transvaginal sonography with color Doppler imaging and color Doppler energy. Pulsed Doppler sonography was used to obtain flow velocity waveforms to determine the pulsatility index (PI), resistance index (RI), peak systolic velocity, and time-averaged maximum velocity. The tumors were classified retrospectively according to histologic criteria.

Results: There were 52 women with benign, three with borderline, and 12 with malignant ovarian tumors. Repeated-measures analysis of variance revealed no systematic differences in the values of all four measurements performed under color Doppler imaging and color Doppler energy for all cases of corpora lutea and adnexal tumors (PI: P=.153, RI: P=.197, peak systolic velocity: P=.355, time-averaged maximum velocity: P=.159). All cases of borderline and malignant tumors had detectable pulsatile blood flow with color Doppler imaging and color Doppler energy. Forty-two (80.8%) of the benign tumors had flow detectable with color Doppler imaging, compared with 40 (76.9%) with color Doppler energy (P=.480). Analysis of receiver operating characteristic curves showed a marginal but nonsignificant improvement in diagnostic performance with color Doppler energy compared with color Doppler imaging for all four measurements (PI: P=.182, RI: P=.178, peak systolic velocity: P=.254, time-averaged maximum velocity: P=.238). The intraclass correlation coefficients for all four measurements were superior with color Doppler imaging compared with color Doppler energy.

Conclusion: Flow velocity waveform analysis and diagnostic accuracy for ovarian malignancy are not significantly different between color Doppler imaging and color Doppler energy. Examinations with color Doppler imaging appear to be more reproducible than those with color Doppler energy.