Magnetic resonance imaging of placentome development in the pregnant Ewe

Abstract

Introduction: Novel imaging measurements of placental development are difficult to validate due to the invasive nature of gold-standard procedures. Animal studies have been important in validation of magnetic resonance imaging (MRI) measurements in invasive preclinical studies, as they allow for controlled experiments and analysis of multiple time-points during pregnancy. This study characterises the longitudinal diffusion and perfusion properties of sheep placentomes using MRI, measurements that are required for future validation studies. Methods: Pregnant ewes were anaesthetised for a MRI session on a 3T scanner. Placental MRI was used to classify placentomes morphologically into three types based on their shape and size at two gestational ages. To validate classification accuracy, placentome type derived from MRI data were compared with placentome categorisation results after delivery. Diffusion-Weighted MRI and T2-relaxometry were used to measure a broad range of biophysical properties of the placentomes. Results: MRI morphological classification results showed consistent gestational age changes in placentome shape, as supported by post-delivery gold standard data. The mean apparent diffusion coefficient was significantly higher at 110 days gestation than at late gestation (~140 days; term, 150 days). Mean T2 was higher at mid gestation (152.2 ± 58.1 ms) compared to late gestation (127.8 ms ± 52.0). Significantly higher perfusion fraction was measured in late gestation placentomes that also had a significantly higher fractional anisotropy when compared to the earlier gestational age. Discussion: We report baseline measurements of techniques common in placental MRI for the sheep placenta. These measurements are essential to support future validation measurements of placental MRI techniques.

Placental MRI Predicts Fetal Oxygenation and Growth Rates in Sheep and Human Pregnancy

“…[ 18 ] Measuring the placenta itself is important for understanding the function of the organ and can be advantageous since placental motion artifacts are less severe compared to the free‐floating moving fetus. [ 24 , 44 , 45 ] However, within any given region of the placenta the measured MRI signal will contain merged information from the independent maternal and fetal blood circulations, thus an alternative model is required when estimating a pure fetal blood T 2 value in this way.…”

“…DTI voxel size was 3.6 × 3.6 × 2.5 mm 3 , reconstructed matrix 78 × 96 and 36 slices. [44] Estimation of Fetal and Brain Weight from MRI Volumetry: A 3D steadystate free procession of the uterus (T E = 1.6 ms; repetition time [TR] = 3.38 ms; slice thickness = 2 mm; number of slices = 100-120; number of averages = 1; matrix size = 272 × 272; field of view (FOV) = 430 × 430 mm; average acquisition time ≈9 min) was acquired and segmented using ITK-SNAP (version 3.8) to measure fetal volume. [71] Overall fetal volume and brain volume were used to estimate both fetal weight and brain weight using a previously described conversion factor.…”

“…[18,27] tissue analysis. [44] The ROIs covered a representative area of each placentome, and ROIs were placed away from edges such that any residual movement artifact would not cause it to move out of the placentome. To reduce motion artifact, a rigid registration [78] was applied followed by a nonrigid free-form registration.…”

“…The placentome regions of interest (ROI) were manually segmented from the first b = 0 image (ITK‐SNAP Version 3.6.0, 2017) and double‐checked by a member of the team familiar with post‐mortem tissue analysis. [ 44 ] The ROIs covered a representative area of each placentome, and ROIs were placed away from edges such that any residual movement artifact would not cause it to move out of the placentome. To reduce motion artifact, a rigid registration [ 78 ] was applied followed by a nonrigid free‐form registration.…”

“…[18] Measuring the placenta itself is important for understanding the function of the organ and can be advantageous since placental motion artifacts are less severe compared to the free-floating moving fetus. [24,44,45] However, within any given region of the placenta the measured MRI signal will contain merged information from the independent maternal and fetal blood circulations, ) and those that are chronically hypoxemic with FGR (< 10 th centile) due to carunclectomy pre-conception leading to placental insufficiency (red triangles, n = 15). Dashed line represents the 10 th centile for control pregnancies (4.0 kg).…”