Evaluation of two different methods for vascular sampling by three-dimensional power Doppler angiography in solid and...

To analyze two different methods for performing three-dimensional power Doppler angiography (3D-PDA) vascular sampling in solid and cystic-solid adnexal masses.

Twenty-one 3D-PDA volumes from 18 consecutive and unselected solid or cystic-solid adnexal masses (13 malignant and five benign) were analyzed. A single examiner (J. L. A.) acquired all the volumes according to a predetermined scanning protocol. Two different observers (one inexperienced and the other experienced) calculated 3D-PDA vascular indices (vascularization index (VI), flow index (FI) and vascularization flow index (VFI)) from solid tumor areas. First, manual sampling (Plane A, 15 degrees rotation-step) was performed, and 1 week later 5-cm(3) sphere sampling from the most vascularized area was carried out. The observers made a record of any difficulty that they encountered in delineating the solid areas of tumors or in distinguishing true tumor vessels from pre-existing vessels, the time spent performing each analysis was recorded and inter- and intraobserver reproducibility was evaluated for each method using intraclass correlation coefficients (ICC).

In four (19.0%) of the 21 volumes sphere sampling could not be performed because it was not possible to obtain a sphere smaller than 5.5 cm(3). This happened in cases in which image zooming was used when acquiring the 3D volume. The inexperienced observer encountered more difficulty, but not significantly more, than the experienced observer when analyzing 3D-PDA volumes both by manual sampling (29% vs. 14% of cases) and 5-cm(3) sphere sampling (35% vs. 18% of cases). The mean time spent by the inexperienced observer was significantly greater (P < 0.001) than that spent by the experienced observer both for manual sampling (6.11 min vs. 1.85 min) and 5-cm(3) sphere sampling (2.93 min vs. 2.15 min). Contrary to the findings for the experienced observer, the inexperienced observer required less time to perform sphere sampling than they did manual sampling. Interobserver agreement was high for both methods: ICC for manual volume, 0.993; manual VI, 0.908; manual FI, 0.913; manual VFI, 0.914; sphere volume, 0.949; sphere VI, 0.954; sphere FI 0.850; and sphere VFI, 0.953. Intraobserver reproducibility was also high, with all ICCs above 0.99.

Manual and 5-cm(3) sphere sampling are reproducible methods for 3D-PDA vascular sampling. Caution is required when image zoom is used at the time of acquiring the volume because this may prevent sphere sampling. Difficulties found in performing both manual and sphere sampling do not seem to significantly affect the reproducibility of Doppler index calculations.