Imaging the human placental microcirculation with micro-focus computed tomography: Optimisation of tissue preparation and image acquisition

Micro-CT provides 3D volume imaging with spatial resolution at the micrometre scale. We investigated the optimal human placenta tissue preparation (contrast agent, perfusion pressure, perfusion location and perfusion vessel) and imaging (energy, target material, exposure time and frames) parameters. Microfil (Flow Tech, Carver, MA) produced better fill than Barium sulphate (84.1%(±11.5%)vs70.4%(±18.02%) p = 0.01). Perfusion via umbilical artery produced better fill than via chorionic vessels (83.8%(±17.7%)vs78.0%(±21.9%), p < 0.05), or via umbilical vein (83.8%(±16.4%)vs69.8%(±20.3%), p < 0.01). Imaging at 50 keV with a molybdenum target produced the best contrast to noise ratio. We propose this method to enable quantification and comparison of the human fetoplacental vascular tree.


Introduction
Fetal health and development is intricately bound with human placental circulation, yet there is no validated quantitative method with which to assess vascularisation of the human placenta. Developing a quantitative method may improve our ability to investigate, and therefore understand, normal placental function and pathologies such as fetal growth restriction, stillbirth and twinto-twin transfusion syndrome.
Micro-focus Computed Tomography (micro-CT) provides threedimensional volume imaging with spatial resolution at the micrometre scale. The technique has been used to investigate the branching structure and tortuosity of the fetoplacental circulation of mouse placentae [1] [2], and shown differences in vascular density of the human placenta between normally grown and growth restricted pregnancies [3], [4].
This study was designed to develop optimised tissue-specific preparation and micro-CT imaging parameters, in order to provide a validated approach to human placenta micro-CT.

Method
This series of experiments is divided into two sections; investigating tissue preparation techniques, and then micro-CT imaging parameters. The full experimental methodology is described in supplementary data.

Tissue acquisition
Experimental procedures were approved by Bloomsbury National Research Ethics Service Committee and by University College London Hospital Research and Development (REC Reference number 133888). Placentas delivered by elective term caesarean section were taken directly to the laboratory, had the membranes trimmed, and the amnion removed.

Tissue preparation comparators
We designed experiments to compare ( The fetal vessel of interest was cannulated, and a cut made in the main draining vessel close to the point of cannulation, to create a fluid exit vent. 0.9% sodium chloride solution with 5IU heparin/ml was perfused until the outflow ran clear, then contrast agent was perfused until the chorionic vasculature was fully perfused and contrast agent was seen in the draining vessel. The vessel was occluded and the contrast agent was left to set. The placenta was dissected into 2 Â 2cm full thickness blocks, which were fixed in 4% formalin for a minimum of 48 h. One full thickness section stained with hematoxylin and eosin (H&E) was cut for every block and 6 micrographs at x100 magnification taken (see supplementary material).
Histological analysis was done in FIJI (ImageJ Version 2.0.0-rc-54/1.51f) [8]. Vascular fill was calculated for each micrograph as shown in equation one.

Micro-CT imaging comparators
We designed experiments to compare (Table 1).
Energy level e from 30 to 100 keV in 10 keV increments. Target material e comparing Tungsten, Copper and Molybdenum. Exposure timee500 and 1000 ms Averaged frames per projectione1 and 2 A 2 Â 2cm full thickness block of human placenta was repeatedly imaged (XT H 225 ST Micro-CT, Nikon Metrology, Tring, UK) adjacent to a 3 mm internal diameter tube filled with Microfil. Scans were reconstructed using a modified Feldkamp filtered back projection algorithm with proprietary software (CTPro3D; Nikon Meterology), and the average greyscale values of recorded areas of interest drawn over placenta, Microfil and air were calculated. The contrast to noise ratio was calculated as shown in equation two. Table 1 Comparison of placental tissue preparation and micro-CT imaging parameters used in this study and in two previous studies, and optimised protocol as determined by the results of this study. SNR ¼ signal to noise ratio.

Statistical analysis
Data is presented as mean ± SD. Statistical analysis was done in SPSS Statistics (IBM version 23). Group comparison was performed using independent sample t-tests with significance set at 95%.

Micro-CT imaging parameters
Contrast and noise were both greatest at the lower energy levels (Fig. 1A/B). The optimal CNR was with Molybdenum target at 50 keV (Fig. 1C). Increasing exposure time from 500 ms to 1000 ms and averaged frames per projection reduced the noise and improved the CNR (Fig. 1D) at the cost of imaging time and throughput (Table 1).

Discussion
We have established optimal tissue and imaging parameters for placental angiographic micro-CT (Table 1). Our studies show that Microfil is a superior contrast agent to barium suphate, and that central and arterial perfusion are superior to peritheral and venous perfusion. Contrast to noise ratio is optimal when imaging with 50 keV energy, with a Molybdenum target. Increasing the number of projection and exposure time improves CNR at the cost of throughput. Our studies found 1000 ms exposure time and 3141 projections over 360 rotation produced good CNR with a 54 min  This approach can be used to investigate the microcirculation of the human placenta. The technique benefits from its high resolution and large field of view, allowing images of the vascular tree to be captured from the chorionic plate to the intermediate villous vessels (see supplementary data for images).
Micro-CT allows measurement of vascular density and analysis of the structure of the vascular trees, which could improve our understanding of the heterogeneity within normal placentae, and the structural changes associated with diseases such as early and late intrauterine growth restriction. This paper presents independent research funded by the National Institute for Health Research (NIHR). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.

Statements of contribution
Rosalind Pratt I declare that I have contributed to the design, acquired the data and performed the analysis of this study, that I am the primary contributor to the manuscript, and that I have seen and approved the final version. I have no conflicts of interest to declare.
J. Ciaran Hutchinson I declare that I have contributed to the design, execution and analysis of this study and that I have seen and approved the final version. I have no conflicts of interest to declare.
Andrew Melbourne I declare that I have contributed to the design, execution and analysis of this study and that I have seen and approved the final version. I have no conflicts of interest to declare.
Maria Zuluaga Valencia I declare that I have contributed to the automated FIJI analysis of histology, and that I have seen and approved the final version. I have no conflicts of interest to declare.
Alex Virasami I declare that I have contributed to histological analysis in this study and that I have seen and approved the final version. I have no conflicts of interest to declare.
Tom Vercauteren I declare that I have contributed to the design, execution and analysis of this study and that I have seen and approved the final version. I have no conflicts of interest to declare.
Sebastien Ourselin I declare that I have contributed to the design, execution and analysis of this study and that I have seen and approved the final version. I have no conflicts of interest to declare.
Neil Sebire I declare that I have contributed to the design, execution and analysis of this study and that I have seen and approved the final version. I have no conflicts of interest to declare.
Owen J Arthurs I declare that I have contributed to the design, execution and analysis of this study and that I have seen and approved the final version. I have no conflicts of interest to declare.
Anna L David I declare that I have contributed to the design, execution and analysis of this study and that I have seen and approved the final version. I have no conflicts of interest to declare.

Conflicts of interest
None.