Something old, something new: digital quantification of uterine vascular remodelling and trophoblast plugging in historical collections provides new insight into adaptation of the utero-placental circulation

Abstract

Abstract STUDY QUESTION What is the physiological extent of vascular remodelling in and trophoblast plugging of the uterine circulation across the first half of pregnancy? SUMMARY ANSWER All levels of the uterine vascular tree (arcuate, radial and spiral arteries (SAs)) dilate ∼2.6- to 4.3-fold between 6 and 20 weeks of gestation, with significant aggregates of trophoblasts persisting in the decidual and myometrial parts of SAs beyond the first trimester. WHAT IS KNOWN ALREADY In early pregnancy, endovascular trophoblasts form ‘plugs’ in the SAs, transiently inhibiting blood flow to the placenta, whilst concurrently the uterine vasculature undergoes significant adaption to facilitate increased blood delivery to the placenta later in gestation. These processes are impaired in pregnancy disorders, but quantitative understanding of the anatomical changes even in normal pregnancy is poor. STUDY DESIGN, SIZE, DURATION Serial sections of normal placentae in situ (n = 22) of 6.1–20.5 weeks of gestation from the Boyd collection and Dixon collection (University of Cambridge, UK) were digitalized using a slide scanner or Axio Imager.A1 microscope. PARTICIPANTS/MATERIALS, SETTING, METHODS Spiral (n = 45), radial (n = 40) and arcuate (n = 39) arteries were manually segmented. Using custom-written scripts for Matlab® software, artery dimensions (Feret diameters; major axes; luminal/wall area) and endovascular trophoblast plug/aggregate (n = 24) porosities were calculated. Diameters of junctional zone SAs within the myometrium (n = 35) were acquired separately using a micrometre and light microscope. Decidual thickness and trophoblast plug depth was measured using ImageJ. MAIN RESULTS AND THE ROLE OF CHANCE By all measures, radial and arcuate artery dimensions progressively increased from 6.1 to 20.5 weeks (P < 0.01). The greatest increase in SA calibre occurred after 12 weeks of gestation. Trophoblast aggregates were found to persist within decidual and myometrial parts of SA lumens beyond the first trimester, and up to 18.5 weeks of gestation, although those present in the second trimester did not appear to prevent the passage of red blood cells to the intervillous space. Trophoblasts forming these aggregates became more compact (decreased in porosity) over gestation, whilst channel size between cells increased (P = 0.01). Decidual thickness decreased linearly over gestation (P = 0.0003), meaning plugs occupied an increasing proportion of the decidua (P = 0.02). LARGE SCALE DATA N/A. LIMITATIONS, REASONS FOR CAUTION Although serial sections were assessed, two-dimensional images cannot completely reflect the three-dimensional properties and connectivity of vessels and plugs/aggregates. Immersion-fixation of the specimens means that vessel size may be under-estimated. WIDER IMPLICATIONS OF THE FINDINGS Uterine vascular remodelling and trophoblast plug dispersion is a progressive phenomenon that is not completed by the end of the first trimester. Our quantitative findings support the concept that radial arteries present a major site of resistance until mid-gestation. Their dimensional increase at 10–12 weeks of gestation may explain the rapid increase in blood flow to the placenta observed by others at ∼13 weeks. Measured properties of trophoblast plugs suggest that they will impact on the resistance, shear stress and nature of blood flow within the utero-placental vasculature until mid-gestation. The presence of channels within plugs will likely lead to high velocity flow streams and thus increase shear stress experienced by the trophoblasts forming the aggregates. Quantitative understanding of utero-placental vascular adaptation gained here will improve in silico modelling of utero-placental haemodynamics and provide new insights into pregnancy disorders, such as fetal growth restriction. STUDY FUNDING/COMPETING INTEREST(S) This work was supported by a Royal Society Te Aparangi Marsden Grant [18-UOA-135]. A.R.C. is supported by a Rutherford Discovery Fellowship [14-UOA-019]. The authors have no conflict of interest to declare.