Assessing repeatability of 3D Doppler indices obtained by static 3D and STIC power Doppler: a combined in-vivo/in-vitro flow phantom study.

OBJECTIVES: To compare the variability in vascularization flow index (VFI) seen in serial acquisitions obtained using spatiotemporal image correlation (STIC) and using conventional static three-dimensional (3D) power Doppler (PD), for both in-vitro and in-vivo models, and to evaluate whether the curves formed by VFI values obtained from successive 'frames' in a STIC dataset are consistent and resemble the waveforms obtained by spectral Doppler analysis. METHODS: The study was divided into two parts: in the first part (the in-vitro model) we scanned a flow phantom, while in the second part (the in-vivo model) we scanned a common carotid artery. Conventional static 3D and STIC-PD datasets were alternately acquired from these two models. VFI values were assessed from 0.38-cm(3) spherical samples of the main flow region in the static 3D datasets and in every frame of the STIC datasets. The variance of the minimum, mean and maximum VFI values from each STIC dataset was compared with the variance of VFI values from the static 3D datasets. RESULTS: Ten static 3D and 10 STIC datasets were acquired from each model. Analysis of the in-vitro and in-vivo models showed a significant reduction in the variance of VFI values obtained using STIC as compared to static datasets. Additionally, we observed that the curves formed by VFI values obtained from successive frames in each STIC dataset were consistent across different datasets and that they resembled the waveforms obtained by spectral Doppler in both models. CONCLUSIONS: 3D-PD indices derived from STIC are more stable than those obtained from conventional static 3D-PD datasets. The curves of VFI throughout a reconstructed cardiac cycle using STIC are repeatable and resemble those obtained by spectral Doppler analysis of the vessel.

Three-dimensional Doppler evaluation of single spherical samples from the placenta: intra- and interobserver reliability.

OBJECTIVE: To evaluate the intra- and interobserver reliability of assessment of three-dimensional power Doppler (3D-PD) indices from single spherical samples of the placenta. METHODS: Women with singleton pregnancies at 24-40 weeks' gestation were included. Three scans were independently performed by two observers; Observer 1 performed the first and third scan, intercalated by the scan of Observer 2. The observers independently analyzed the 3D-PD datasets that they had previously acquired using four different methods, each using a spherical sample: random sample extending from basal to chorionic plate; random sample with 2 cm(3) of volume; directed sample to the region subjectively determined as containing more color Doppler signals extending from basal to chorionic plate; or directed sample with 2 cm(3) of volume. The vascularization index (VI), flow index (FI) and vascularization flow index (VFI) were evaluated in each case. The observers were blinded to their own and each other's results. Additional evaluation was performed according to placental location: anterior, posterior and fundal or lateral. Intra- and interobserver reliability was assessed by intraclass correlation coefficients (ICC). RESULTS: Ninety-five pregnancies were included in the analysis. All three placental 3D-PD indices showed only weak to moderate reliability (ICC < 0.66 and ICC < 0.48, intra- and interobserver, respectively). The highest values of ICC were observed when using directed spherical samples from basal to chorionic plate. When analyzed by placental location, we found lower ICCs for lateral and fundal placentae compared to anterior and posterior ones. CONCLUSION: Intra- and interobserver reliability of assessment of placental 3D-PD indices from single spherical samples in pregnant women greater than 24 weeks' gestation is poor to moderate, and clinical usefulness of these indices is likely to be limited.

Biometry and fetal weight estimation by two-dimensional and three-dimensional ultrasonography: an intraobserver and interobserver reliability and agreement study.

OBJECTIVE: To evaluate and compare the intraobserver and interobserver reliability and agreement for the biparietal diameter (BPD), abdominal circumference (AC), femur length (FL) and estimated fetal weight (EFW) obtained by two-dimensional ultrasound (2D-US) and three-dimensional ultrasound (3D-US). METHODS: Singleton pregnant women between 24 and 40 weeks were invited to participate in this study. They were examined using 2D-US in a blinded manner, twice by one observer, intercalated by a scan by a second observer, to determine BPD, AC and FL. In each of the three examinations, three 3D-US datasets (head, abdomen and thigh) were acquired for measurements of the same parameters. We determined EFW using Hadlock's formula. Systematic errors between 3D-US and 2D-US were examined using the paired t-test. Reliability and agreement were assessed by intraclass correlation coefficients (ICCs), limits of agreement (LoA), SD of differences and proportion of differences below arbitrary points. RESULTS: We evaluated 102 singleton pregnancies. No significant systematic error between 2D-US and 3D-US was observed. The ICC values were higher for 3D-US in both intra- and interobserver evaluations; however, only for FL was there no overlap in the 95% CI. The LoA values were wider for 2D-US, suggesting that random errors were smaller when using 3D-US. Additionally, we observed that the SD values determined from 3D-US differences were smaller than those obtained for 2D-US. Higher proportions of differences were below the arbitrarily defined cut-off points when using 3D-US. CONCLUSION: 3D-US improved the reliability and agreement of fetal measurements and EFW compared with 2D-US.