Micro-CT and histological investigation of the spatial pattern of feto-placental vascular density

Abstract

Introduction: There are considerable variations in villous morphology within a normal placenta. However, whether there is a reproducible spatial pattern of variation in villous vascular density is not known. Micro-CT provides three-dimensional volume imaging with spatial resolution down to the micrometre scale. In this study, we applied Micro-CT and histological analysis to investigate the degree of heterogeneity of vascularisation within the placenta.

Method: Ten term placentas were collected at elective caesarean section, perfused with contrast agent and imaged whole with Micro-CT. Eight full depth tissue blocks were then taken from each placenta and imaged. Sections were taken for histological analysis. Data was analysed to investigate vascular fill, and vascular density in relation to location from cord insertion to placental edge at each scale.

Results: Whole placental imaging revealed no spatially consistent difference in villous vessel density within the main placental tissue, although there was a great degree of heterogeneity. Both block imaging and histological analysis found a large degree of heterogeneity of vascular density within placentas, but no strong correlation between villous vascular density and block location (r_s = 0.066, p = 0.7 block imaging, r_s = 0.06, p = 0.6 histological analysis).

Discussion: This work presents a novel method for imaging the human placenta vascular tree using multiscale Micro-CT imaging. It demonstrates that there is a large degree of variation in vascular density throughout normal term human placentas. The three-dimensional data created by this technique could be used, with more advanced computer analysis, to further investigate the structure of the vascular tree.

Fig. 1

Micro-CT imaging of a human placenta perfused with Microfil. Surface renderings made using VG StudioMAX 2.2 (Volume Graphics, Germany) were thresholded halfway between the grey scale intensities of tissue and Microfil. A and B; the whole placenta, imaged with an isotropic voxel size of 116.5 μm. C; a slice through the whole placenta, showing the geometric arrangement of chorionic and villous vessels. D and E; two blocks imaged with an isotropic voxel size of 13.5 μm. The complex vascular tree is clearly seen, with whole imaging showing chorionic and stem vessels, and block imaging showing the villous vascular tree down to the terminal capillaries.

Fig. 2

FIJI histological analysis pipeline.

Fig. 3

A and B; example of normalised distance maps radiating out from the umbilical cord insertion for placenta 3 (A) and 9 (B), C and D; example vascular density maps for the same placenta, E and F; graphs showing mean vascular density for each of the 100 regions from the umbilical cord insertion (0) to the placental edge (100), for each placenta. G; the combined mean vascular density with distance from the umbilical cord insertion with error bars showing standard deviation.

Fig. 4

A and B: Box plots showing the spread of block vascular density (box shows 25th to 75th centile, with midline showing the median) between placentas, measured with block μCT (A) and histological analysis (B). C and D; Graphs showing correlation between block villous vascular density and normalised block location in relation to the umbilical cord insertion (0) and placental edge (100) measured with block μCT (C) and histological analysis (D).