Cervical blood perfusion assessed using power Doppler‐derived estimation of fractional moving blood volume: a reproducibility study

Abstract

To evaluate the reproducibility of estimation of power Doppler-derived fractional moving blood volume (FMBV) in the uterine cervix of women with uncomplicated pregnancies. Two experienced operators evaluated 30 uncomplicated singleton pregnancies at 20-24 weeks of gestation. The PDU box was positioned in a mid-sagittal view of the cervix, including the internal os and external os, in the same plane as that in which cervical length is measured. Two consecutive examinations were performed by each operator, in each of which the cervical length was measured and five consecutive good-quality images with PDU information were obtained. The region of interest (ROI) (cervix) was delineated offline and FMBV, which expresses the percentage of blood occupying the ROI, was calculated with purpose-designed software. Intra- and interobserver intraclass correlation coefficients (ICCs) and mean differences with 95% limits of agreement (LOA) were calculated. The median gestational age at examination was 22 + 0 weeks. Measurements (median ± SD) obtained for cervical length and FMBV were as follows: Operator A: 37 ± 7.4 mm and 8.11 ± 2.9%; Operator B: 37.5 ± 9.3 mm and 7.9 ± 3.3%, respectively. The intra- and interobserver ICCs for FMBV were 0.88 (95% CI, 0.75-0.94) and 0.82 (95% CI, 0.64-0.94), respectively. There was a mean difference in FMBV measurement between operators of - 0.2 ± 1.75% (95% LOA, - 3.7 to 3.2%). Reproducible assessment of cervical blood perfusion through estimation of FMBV can be achieved while cervical length is being measured.