Transvaginal color Doppler assessment of venous flow in adnexal masses.

OBJECTIVE To analyze the usefulness of transvaginal color Doppler assessment of venous flow in the differential diagnosis of adnexal masses. MATERIAL AND METHODS Ninety-one consecutive patients (mean age: 46.6 years, range: 16-81 years) diagnosed as having an adnexal mass were evaluated by transvaginal color Doppler sonography prior to surgery. Color Doppler was used to detect and analyze the flow velocity waveform from arterial and venous blood flow within the tumor. For arterial signals the resistance index and peak systolic velocity, and for veins the maximum venous flow velocity, were calculated. Receiver operator characteristic curves were plotted to determine the best venous flow velocity cut-off. According to our previous study using arterial Doppler, a tumor was considered as malignant when flow was detected and the lowest resistance index was < or = 0.45. Using venous Doppler a mass was considered as malignant when flow was detected and the venous flow velocity was > or = the best cut-off found on the receiver operator characteristic curve. Definitive histopathological diagnosis was obtained in all cases. Sensitivity, specificity, positive predictive value and negative predictive value for B-mode morphology (evaluation performed according to Sassone's scoring system), arterial Doppler, venous Doppler, and a combination of both arterial and venous Doppler were calculated. RESULTS Twenty-five masses (27.5%) were malignant and 66 (72.5%) benign. Arterial and venous flow was found more frequently in malignant than in benign masses (92% vs. 41% (P < 0.001) and 72% vs. 21% (P < 0.001), respectively). The resistance index was significantly lower in malignant tumors (0.42 vs. 0.60, P = 0.0003). No differences were found in peak systolic velocity. Venous flow velocity was significantly higher in malignant masses (18.1 cm/s vs. 8.9 cm/s, P = 0.0006). The best cut-off of venous flow velocity was 10 cm/s. Sensitivity, specificity, positive predictive value and negative predictive value for morphology, arterial Doppler, venous Doppler, and the combination of both arterial and venous Doppler were 92%, 71%, 45%, 96%; 76%, 95%, 87%, 91%; 68%, 94%, 81%, 89%; and 88%, 91%, 79%, 95%, respectively. CONCLUSIONS Our results indicate that preoperative evaluation by venous flow assessment of adnexal masses may be useful to discriminate between malignant and benign tumors.