Three-dimensional power Doppler sonography: imaging and quantifying blood flow and vascularization

Abstract

Objectives: To assess the feasibility of imaging low-velocity blood flow in adnexal masses by transvaginal three-dimensional power Doppler sonography, to analyze three-dimensional power Doppler sonography data sets with a new computer-assisted method and to test the reproducibility of the technique.

Methods: A commercially available 5-MHz Combison 530 ultrasound system was used to perform three-dimensional power Doppler sonography transvaginally. A cube (= volume of interest) was defined enclosing the vessels of the cyst and the Cartesian characteristics were stored on a hard disk. This cube was analyzed using specially designed software. Five indices representing vascularization (the vascularization index (VI) or blood flow (the flow index (FI)) or both (the vascularization-flow index (VFI)) were calculated. The intraobserver repeatability of cube definition and scan repetition was assessed using Hartley's test for homogeneous variances. Interobserver agreement was assessed by the Pearson correlation coefficient.

Results: Imaging of vessels with low-velocity blood flow by three-dimensional power Doppler sonography and cube definition was possible in all adnexal massed studied. In some cases even induced non-vascular flow related to endometriosis was detected. The calculated F value with intraobserver repeated Cartesian file-saving ranged from 0 to 18.8, with intraobserver scan repetition from 4.74 to 24.8 for VI, FI 1, FI 2 and VFI 1; for VFI 2 the calculated F value was 64. The interobserver correlation coefficient ranged between 0.83 and 0.92 for VI, FI 1, FI 2 and VFI 1; for VFI 2 the correlation coefficient was less than 0.75.

Conclusion: Vessels with low-velocity blood flow can be imaged using three-dimensional power Doppler sonography. Induced non-vascular flow was detected in endometriotic cyst fluid. Three-dimensional power Doppler sonography combined with the cube method gave reproducible information for all indices except VFI 2. These indices might prove to be a new predictor in all fields of neoangiogenesis. The clinical relevance remains to be determined.