All content on this website, including dictionary, thesaurus, literature, geography, and other reference data is for informational purposes only. This information should not be considered complete, up to date, and is not intended to be used in place of a visit, consultation, or advice of a legal, medical, or any other professional.
The spiral arteries are remodeled, and with the absent muscle layer, they are widely patent and allow more flow to the placenta and the flow is also not able to be controlled by maternal vascular tone. The arteries are remodeled quite deep in the uterus and so the sampled area seen attached to the decidua is not necessarily adequately represent the degree of remodeling. The remodeling is probably a continuous process rather than one that occurs in two waves. The other effect of widely patent spiral arteries is that it produces lower flow into the intervillous space and causes less shear damage to the overlying trophoblast. This may be a factor in the increased trophoblastic debris seen in preeclampsia where high-pressure jets of maternal blood emanate from the narrowed spiral arteries.
Atherosis is a process that is not well understood. It consists of foamy macrophages with fibrinoid change often seen best in the decidua at the margin of the placenta. Histologically, it is reminiscent of acute humoral rejection in transplants, hemolytic uremic syndrome, and, also, cyclosporin A toxicity, all of which are related to endothelial toxicity. There may be a correlation with long-term cardiovascular disease.
A specific population of extravillous trophoblasts invades towards uterine spiral arteries, percolates the walls of these vessels, displaces the endothelial lining and finally reaches the lumen of the arteries (Kaufmann etal., 2003). This endoarterial trophoblast erodes, opens and transforms uterine spiral arteries to enable hemotrophic nutrition during the second and third trimester of pregnancy.
The main function of endoarterial trophoblasts is not only opening of the spiral arteries but most of all their transformation into large conduits that have lost the maternal control on their contractility. It needs to be clarified that widening of these vessels takes place at the very end of the arteries towards the placenta.
Maternal factors induce first changes of the uterine spiral arteries very early in pregnancy prior to trophoblasts reaching such vessels. The vessels show reduced organization of vascular smooth muscle cells, altered morphology of endothelial cells and first signs of dilation. It seems as if maternal immune cells trigger these early changes. They accumulate within the decidua in close vicinity to spiral arteries and play an active role in transforming these vessels (Benirschke etal., 2006).
After disorganization of the vessel wall and first slight dilation of the vessel, endoarterial trophoblasts percolate from the decidual stroma into the vessel wall. During the very first invasion endoarterial trophoblasts reach the lumen and form plugs to block blood flow towards the placenta (Weiss etal., 2016).
Later, endoarterial trophoblasts infiltrate deeper regions of the vessels up to the myometrial portion of the arteries. This results in a further dilation of the vessels, finally reaching a multiple of the original diameter. The reduced activity of smooth muscle cells plus the loss of elastic fibers in the vessel wall further contribute to the dilation effect (Benirschke etal., 2006).
During early pregnancy, the spiral arteries undergo extensive remodeling. This remodeling is driven by extravillous trophoblast cells as described earlier, as well as by uNKs.58 uNKs are evolutionarily conserved, having been shown to accumulate in the uterus during pregnancy in species including rodent, pig, and human.59 uNK cells are recruited to the uterus via local chemokine actions. For instance, trophoblasts secrete CXCL12 and endometrial cells secrete CXCL10 and CXCL11 to attract uNK cells. uNK cells bind these chemokines with receptors including CXCR3 and CXCR4.60 Recruitment of uNK cells also involves the chemokine CXCL14, as evidenced by the fact that Cxcl14/ mice have significantly decreased uNK cells in the uterus during pregnancy.61
uNK cells can cause trophoblast-independent spiral artery remodeling and facilitate the invasion of extravillous trophoblasts. uNK cells play a role in spiral artery remodeling by secreting factors such as interferon-gamma (IFNG).62 IFNG finely regulates the migration of extravillous trophoblasts.63 IFNG also regulates gene expression in human uterine microvasculature endothelial cells, inducing genes involved in angiogenesis and uNK cell recruitment.64 In addition, estrogen induces uNK cells to produce the chemokine CCL2,65 which is another known angiogenic factor.66 In addition to spiral artery remodeling, uNK cells induce angiogenesis and increase microvasculature density in part via expression of VEGFA.67
While uNK cells are critical for remodeling spiral arteries and inducing angiogenesis during early pregnancy, these cells must be closely regulated to prevent embryonic death. Increased uNK cell numbers are associated with recurrent reproductive failure, possibly due to early onset of maternal blood flow leading to fetal oxidative stress.68 Similarly, uNK cell numbers are dysregulated during the menstrual cycle of women with heavy menstrual bleeding.69
Trophoblast invasion and remodeling of the spiral arteries is in part regulated by integrins and other adhesion molecules. Cytotrophoblasts initially express epithelial cell-type adhesion molecules such as integrins 6/4 and 6/1, and E-cadherin. During normal pregnancy cytotrophoblasts become more invasive, and the epithelial cell-type adhesion molecules are replaced by the endothelial-type integrins 1/1 and V/3, a process known as vascular mimicry or pseudovasculogenesis (Fig. 2).84 These phenotypic changes in integrins may be impaired during placental hypoxia and preeclampsia. Hypoxia alters the placental expression of integrins and fibronectin, causing increased expression of integrin 5 and fibronectin and decreased expression of integrin 1.85 Also, during preeclampsia, abnormal expression of epithelial cell-type adhesion molecules and apoptosis of cytotrophoblasts cause limited invasion of spiral arteries, placental ischemia, and RUPP.84,86,87 Ezrin is one of the integrins involved in cell adhesion, organization, and migration. Ezrin is downregulated in syncytiotrophoblast microvesicles from preeclamptic women, resulting in reduced invasiveness of cytotrophoblats, shallow placentation, and defective vascularization of the placenta.88 The decreased trophoblast invasion and replacement of vascular cells also lead to retention of VSM cells in the spiral arteries, which promote vasoconstriction,89 and further decrease uteroplacental blood flow and aggravate placental ischemia (Fig. 2).
Endothelial cell adhesion molecules such as soluble intercellular adhesion molecule-1 (ICAM-1) and soluble vascular cell adhesion molecule-1 (VCAM-1) are downregulated during normal pregnancy, thus minimizing leukocyte adhesion to endothelial cells and maintaining patency and blood flow in the spiral arteries. The plasma levels of ICAM-1 and VCAM-1 are increased in preeclampsia, leading to increased leukocyte adhesion to endothelial cells and restricted blood flow in the spiral arteries.90
Preeclampsia is also associated with increased placental expression of microRNA miRNA-125b-1-3p which could reduce the expression of S1PR1, a G-protein-coupled receptor that facilitates invasion of human trophoblasts.91 Preeclampsia is also associated with increased expression of placental miRNA-517a/b and miRNA-517c, which have been shown to be expressed and to decrease trophoblast invasion in extravillous trophoblasts under hypoxic conditions 92.
The uterus contains 46000 spiral arteries with approximately 150 of them contributing to the placental bed, perfusing the intervillous space. Trophoblastic invasion of these maternal spiral arteries is a key factor in proper placental implantation and promotion of healthy placental vascular growth. There are two distinct stages, trophoblastic plugging and invasive arterial remodeling.
The former contributes to the early placental hypoxic environment, which promotes placental vasculogenesis and branching angiogenesis, while the latter is important for meeting the oxygen demands of the developing fetus and elongation of the placental vascular system through nonbranching angiogenesis.
As presented in Chapter 2, during implantation, anchoring villi form polarized trophoblastic columns, from which extravillous trophoblasts detach (Figure 6). These extravillous trophoblasts invade the maternal decidua and distal ends of the spiral arteries. Via complex interactions likely involving maternal decidua, glands, and immune cells, the interstitial trophoblasts migrate toward the spiral arteries, invade the arteries, and become endovascular trophoblasts. The contribution of intravasation from the tissue surrounding the vessel into and through the vessel wall and extravasation with cell starting in the vascular lumen is unclear, but the former seems to predominate. Early in gestation, the trophoblasts accumulate in the vessel lumen creating occlusive trophoblastic plugs (Figure 7(a) and 7(b)). This is followed by switching to an invasive trophoblastic phenotype in which the trophoblasts invade and, along with fibrin, replace the vascular walls that is accompanied by apoptosis of the native endothelium and degradation of the vascular wall components (Figure 7(c)). The result is a high-flow, low-resistance maternalplacental circulation.
Figure 7. (a) Normal spiral artery lined by endothelium and smooth muscle of the media. (b) First trimester is marked by plugging of the spiral arteries by extravillous trophoblasts resulting in a relatively hypoxic environment for placental vascular development. (c) Transformed spiral artery lined by extravillous trophoblasts resulting in fixed vascular dilation and increased maternal blood flow.
Plugging seems to be more prominent in the arteries underneath the center of the placenta with relative sparing of the periphery and accordingly more complete remodeling of these central vessels leads to increased vascular flow compared to the periphery later in gestation. Decidua plays an important role in preventing excessively deep invasion of the trophoblasts into the maternal circulation. This process also takes place in the venous side of the maternal circulation but to a much lesser extent.
Blood flow velocity through the uterine spiral arteries is relatively constant during the second half of pregnancy with values between 33cms1 and 50cms1 (Bahlmann etal. 2012). Burton etal. (2009) have used assumption on length and diameter of vessels, total blood volume and other measures to calculate the velocity of blood flowing into the placenta. The funnel-shaped opening of the vessels reduces the flow of maternal blood into the placenta to values of about 10cms1 while it does not significantly change the total blood volume flowing into the placenta (Burton etal. 2009). This low velocity is essential to maintain the fragile construction of the villous trees, to allow distribution of blood through the narrow passages between the floating villi and to maintain the connection between anchoring villi and basal plate.
A characteristic lesion in placental bed pathology is acute atherosis although it is best observed in the maternal vessels in the decidua parietalis (decidua attached to the extraplacental membranes) since it affects the maternal vessels that have not undergone physiological transformation. It is generally agreed that acute atherosis can be seen in preeclampsia, hypertensive disease not complicated by preeclampsia, normotensive intrauterine growth restriction, and systemic lupus erythematosus, but its presence in uncomplicated diabetes mellitus or gestational diabetes is disputed. The incidence of acute atherosis ranges from 41% to 48% in a series examining placental bed biopsies, placental basal plates, and amniochorial membranes in preeclampsia and intrauterine growth restriction. The correlation between acute atherosis and birth weight, degree of proteinuria and severity, or duration of the hypertension is unclear.
Initially, there is fibrinoid necrosis of the arterial wall and, in established cases, a perivascular lymphocytic infiltrate and lipid-laden macrophages within the lumen and the damaged vessel wall are seen additionally (Figure 7). Another reason that acute atherosis is better seen in the decidua parietalis is that the fibrinoid matrix in the physiologically transformed uteroplacental arteries set in the abundant fibrinoid material within the Rohr's and Nitabuch layer in the basal plate can be mistaken for the fibrinoid necrosis. Acute atherosis has two unwelcome consequences. It results in endothelial disruption that predisposes the vessel to thrombosis while the weakening of the vessel wall can result in aneurysmal formation.
Another less well-recognized lesion is the finding of endovascular trophoblast within the lumina of the spiral arteries in the third trimester in preeclampsia or intrauterine growth restriction (Figure 8). In the first and second trimesters, their presence is part of the vascular response to pregnancy and their absence then is associated with the absence of the physiological vascular changes. In the third trimester, however, the presence of endovascular trophoblast within the lumina of the spiral arteries is pathological and is viewed as a delayed teleological response to the absent vascular changes. Disruption of the vascular endothelium by the endovascular trophoblast adds to the thrombogenicity of the vessel. It may impede blood flow in the same way that intravascular plugging by endovascular trophoblast reduces blood flow into the intervillous space in the first trimester.
A word about the pathological examination of the maternal blood supply is relevant. Since the spiral arteries terminate as openings into the intervillous space in the basal plate of the placenta, maternal vasculopathy in the form of absent physiological change in the decidual segments, intraluminal endovascular trophoblast, thrombosis, or acute atherosis can be detected in sections of the basal plate. Slices of the basal plate taken parallel to the maternal surface, as en face sections, increase the frequency of observing these lesions (Figure 9). Similarly, in the evaluation of acute atherosis in the amniochorial membranes, stacked slices of the membranes have maximized detection rates.
Figure 9. En face sampling (green interrupted lines) can potentially examine more spiral or uteroplacental arteries and their pathology in the basal plate than conventional full-thickness block sampling (black interrupted lines).
The triggers for abnormal cytotrophoblast differentiation and migration toward the spiral arteries are not completely understood. Recent work implicates altered oxygen concentrations at the maternalfetal interface (Figure 535). When villous explants are cultured under hypoxic conditions, they are induced to proliferate (47,79); however, assays of differentiation that rely on cell surface protein expression suggests a failure to differentiate (51,52). The proliferative response to hypoxic conditions may be mediated by the transcription factor Hif-1. Hif-1 has a and b subunits. It is the a subunit that is upregulated during hypoxic conditions, whereas, under either normoxic or hyperoxic conditions, the degradation of this subunit occurs (102). Within placental tissue, Hif-1 was shown to be upregulated during conditions of hypoxia within the placental microenvironment in in vivo and in vitro experiments (123,150). Hif-1 also appears to regulate transcription of TGF-b 3 (131). Strong evidence suggests TGF-b 3 inhibits differentiation and endovascular invasion by cytotrophoblasts (24,25). Thus, Hif-1 may be a key regulator of gene expression important in trophoblast differentiation and migration.
FIGURE 53-5. Hypoxia within the microenvironment of implantation may be responsible for abnormal placentation observed in preeclampsia. The -subunit of the transcription factor Hif-1 is under control of oxygen sensors and ubiquitinated under normoxic or hyperoxic condition. Solid line, hypoxia; dashed line, normoxia.
Early in normal placental development, extravillous cytotrophoblasts invade the uterine spiral arteries of the decidua and myometrium. These invasive fetal cells replace the endothelial layer of the uterine vessels, transforming them from small resistance vessels to flaccid, high-caliber capacitance vessels.31,32 This vascular transformation allows the increase in uterine blood flow needed to sustain the fetus through the pregnancy (Fig. 25-1).
In preeclampsia, this transformation is incomplete (see Fig. 25-1).33,34 Cytotrophoblast invasion of the arteries is limited to the superficial decidua, and the myometrial segments remain narrow and undilated.35,36 Fisher and colleagues have shown that in normal placental development, invasive cytotrophoblasts downregulate the expression of adhesion molecules characteristic of their epithelial cell origin and adopt an endothelial cell surface adhesion phenotype, a process referred to as pseudovasculogenesis.37,38 In preeclampsia, cytotrophoblasts do not undergo this switching of cell surface integrins and adhesion molecules39 and fail to adequately invade the myometrial spiral arteries. The factors that regulate this process are just beginning to be elucidated. Invasive cytotrophoblasts express several angiogenic factors and receptors, including vascular endothelial growth factor (VEGF), placental growth factor (PlGF), and VEGFR-1 (Flt1); expression of these proteins by immunolocalization is altered in preeclampsia.40 Liu and colleagues found that expression of CD146, an endothelial adhesion molecule normally expressed by invasive/migratory cytotrophoblasts, was absent in preeclampsia.41 More work is needed to uncover the molecular signals governing cytotrophoblast invasion early in placentation. These mechanisms are sure to hold key insights into the pathogenesis of preeclampsia.
Soon after implantation in human, rapidly growing cytotrophoblast occludes the superficial spiral arteries to that maternal blood cannot flow into the developing intervillous space. During this period (the first 11weeks, coinciding with organogenesis) the placenta is very hypoxic, with an oxygen partial pressure of<20mmHg at 8weeks. By 12weeks it has risen to >50mmHg and protective antioxidant enzymes, such as catalases, GPXs and SODs have begun to be expressed in the placenta. Antioxidant enzymes in decidual cells offer protection for the developing placenta.
Meanwhile trophoblast invades more deeply, both into the decidual stroma and arteriolar walls. Changes to the arterioles are initiated by maternal uterine NK cells and macrophages, with disruption of the smooth muscle layers and endothelial hypertrophy, and the invasive trophoblast population collaborates in this process as it moves deeper through the decidua. Vascular smooth muscle is eventually replaced by trophoblast embedded in a fibrinoid extracellular matrix, so that the ability to respond to vasoconstrictors is reduced or lost.
At 11weeks the trophoblast plugs are displaced and maternal blood can flow through the intervillous space. Thus the placental interface becomes hemochorial rather than deciduochorial. Trophoblast invasion progresses more deeply into the endometrium, crossing the junctional zone and continuing as far as the inner third of the myometrium. Spiral arterial segments are transformed to the corresponding level.
The idea that decidua may acts to restrain the invasion of trophoblast has gained traction from the genetics of imprinting. Imprinted genes are differentially expressed from their maternal and paternal alleles, a phenomenon found only in eutherian species. An example that has been explored in mouse is igf2 (encoding insulin-like growth factor II), which is important in stimulating fetal growth. Igf2 is expressed exclusively from the paternal allele, and imprinting theory interprets this as a paternal drive to produce larger offspring, which must be constrained by the mother so that her resources are not invested disproportionately in this offspring as compared with those arising in other pregnancies, and her continuing survival is secured. In turn, the gene encoding the recycling receptor igfr2 is imprinted with expression from the maternal allele, so the amount of available IGF is influenced by imprinting from the maternal side. Whether this applies to the uterine control of invasive trophoblast is unclear. However there is evidence in mouse that trophoblast does influence decidual gene expression: antimesometrial decidua differs in its gene expression signature from mesometrial decidua, arguably because the two are receiving paracrine signals from different populations of placentally-derived trophoblast.
For the baby to thrive, the mother must provide the placenta with a large blood flow from her uterine circulation. The increase in uterine blood flow to the placenta has been estimated at 8 to 10 times baseline flow. The body can increase blood flow to a local tissue by relaxing the muscles around arterioles (small arteries), for example when an inflamed area becomes red or when a athlete needs to shunt more blood to active muscles. This strategy works because resistance to blood flow is lowered in this segment of the circulation compared to other areas. Unfortunately this usual strategy of widening arterioles is not able to provide enough increased blood flow for the placenta. A unique strategy is required that is specific for pregnancy and the placenta. This process can go wrong. If it does the fetus may not get enough nutrients or oxygen. The mother may develop preeclampsia. The placenta must adapt to decreased blood flow, and may suffer placenta infarctions and even premature placental separation. This chapter will focus on the normal and pathologic changes in these uterine vessels.
The basic physical principle is that blood flow is directly related to blood pressure (push harder and more will flow) and indirectly to resistance (the less force needed, the more blood will flow). The resistance is fundamentally due to friction against the blood vessel wall, the wider and shorter the blood vessel, the less the resistance to flow. Poiseuilles equation relates the flow of a Newtonian (non-turbulent) fluid in a tube, and although blood flow in a vessel may not meet all the conditions of the equation, it provides a good approximation for understanding the basic parameters governing blood flow.
Q is the blood flow. P is the pressure change, which in the closed circulation is the difference between the arterial and the venous pressure. R is the radius of the blood vessel. The wider the vessel, the less of the flowing blood is rubbing against the vessel wall. The volume passing through is related to the area of the cross section r2, while the wall circumference increases linearly r. This relationship will apply for the whole length of the blood flow (L). The longer the vessel, the more resistance. The more viscous the blood (), the more resistance. If the pressure gradient, and the length of the vessel and viscosity are all kept constant, increasing the radius of the vessel by 2 should increase the flow by a factor of 16!
A larger radius will also have an effect on the flow velocity. For the same pressure change, if more volume is being moved, the velocity will slow down because of conservation of momentum. My understanding is that this is a closed circuit, and the mass is at a certain velocity in the lower radial artery, as the vessel widens, the volume becomes larger, and in order to keep the same mass moving through the system, this mass must be moving at a slower speed. This slower flow will aid gas and nutrient transfer between mother and fetus.
The standard picture of the bodys circulation is a symmetrical pattern of dichotomous branching in which each vessel branches into to two smaller vessels until the capillaries are reached, and then the process reverses going with dichotomous fusing of veins into progressively larger vessels. This fractal pattern holds more or less in most organs, except for collateral vessels that branch across the pattern to provide an alternative route of blood.
The arterial pattern in the uterus follows the expected pattern from the main uterine arteries that branch dichotomously through the outer connective tissue (serosa) and the muscular body (myometrium). There are radial arteries that run perpendicularly across the myometrium, somewhat anomalously to the usual pattern, that are critical to the development of the placental circulation, especially in the endometrium. This endometrial circulation because of menstrual shedding and pregnancy has some unique attributes.
My understanding of the uterine circulation was based on reading the studies of the endometrial circulation in the monkey published by Elizabeth Ramsey. She and other have cited critical earlier studies, but her monograph on the topic is a lucid and insightful summary. Her paper of her injection studies in the original Carnegie Institute Contributions to Embryology paper shows the frustrations and painstaking effort of those studies1. Her monograph on the subject with Martin Donner Placental Vasculature and Circulation details her impressive radiologic injections as well as the anatomic injection studies 2.
I will briefly relate my working understanding of the monograph. Her work was done using monkeys as subjects. They differ from humans in a few details, such as not developing swollen decidual cells in the endometrium and forming a secondary placental lobe on the opposite wall of the uterus. Perhaps most importantly, the human placenta over the pregnancy expands the base of involved arteries tapped for the placenta beyond those of the implantation site, unlike other primates. The rhesus monkey typically utilizes 8 to 12 arteries compared to 120 to 320 in humans. (I have trouble understanding this because studies of essentially cotyledons of spiral artery flow into the placenta number 30-40. This requires some kind of consolidation of groups of spiral arteries.) Other differences include the thinning of the human endometrium and the marked expansion of blood flow beyond that in the primate. At the end of pregnancy the uterine vessels that pre-pregnancy carried a few milliliters of blood per minute, carry approximately 600 ml/min.
Anatomic injection studies show that the radial arteries are continuous with the spiral arteries of the endometrium, and there is in the base of the endometrium basal arties that extend horizontally from the base of the radial arteries and are not modified by pregnancy. The spiral arteries undergo progressive dilatation throughout pregnancy, a process that extends into the myometrium (Fig 1).
The average diameter of the spiral arteries before pregnancy is 200 microns, and this enlarges to 500 microns at term. The basal arteries in the endometrium remain at an average diameter of 180 microns3.
Fig 2: This is an illustration from Drs. Ramsey and Donners monograph of radiologic injection into the uterine circulation of the pregnancy monkey. The entire placenta is not circulated but rather separate spiral arteries in this one injection sequence.
Even in mid-pregnancy , the weak uterine contractions inhibited uterine blood flow. During labor contractions, blood remained in the intervillous space providing some gas exchange even during the cessation of flow due to uterine contraction.
Other prominent students of the utero-placental anatomy include Maurice Panigel4, and the Hamilton and Boyd monograph on the human placenta5, among many others. Some of the history of our understanding of this anatomy is cited in Ramsey and Donor monograph and in a 2006 summary of our knowledge of the spiral arteries by Pijnenborg and colleagues6.
There are two potential ways that a blood vessel could widen. First it could simply grow bigger by proliferating the muscle and connective tissue cells of the wall. This is the way vessels normally respond to increased blood flow as organs grow. The increased flow increases pushes the vessel wall outward. This wall tension is counteracted by cell proliferation increasing the diameter and thickness of the vessel to a larger sustainable size for the flow and pressure.
The second possible way that a blood vessel may widen is that a weakened wall simply expands from the wall tension until a new equilibrium is reached, or the wall ruptures. The latter is occurs often with devastating effect in the focal wall weakness that produces an aneurysm. In pregnancy, the arterial vessels supplying the placenta are widened by destruction of the vascular media in a circumferential process. The possibility of progressive aneurysmal dilatation as an occasional cause of thrombosis or of vascular rupture and hemorrhage cannot be excluded. The coordination of the loss of the media of the vessel involves extravillous trophoblast invasion outside the vessels as well as intravascular trophoblast undermining of endothelial cells. Initially, the vascular lumen may be plugged with trophoblast. At the conclusion of the remodeling the vessel is widened with its wall replaced by fibrinoid and intravascular cytotrophoblast. Numerous biochemical communications are involved including signals from endothelial cells and inflammatory cells. A coordinated quantitative hypothesis of these signals in space with the integration of cell responses does not yet exist6.
The pathologist may observe spiral arteries in curettage samples from spontaneous abortions as well as in the decidualized endometrium attached to the delivered placenta and membranes. These are limited samples either because they are not oriented from the curettage or partial vessels from the separation of the decidua. Less frequently, direct, oriented visualization of deeper spiral and radial arteries is possible from uteri of gravid hysterectomies or autopsies. These chance observations of spiral arteries, as well human placental bed biopsies taken in the context of research, have informed the development of pathologic observations and diagnoses.
I intend to ask colleagues to discuss the large volume of papers on the subject of placental bed biopsies. Much of how we interpret changes in other decidual and myometrial samples is informed by these studies.
The pathologist can evaluate the adequacy of utero-placental flow by the adaptations of placental villi, a topic that will be considered in a later section. The role of utero placental perfusion in intrauterine growth restriction is potentially another measure of trophoblast vascular remodeling that will also be discussed in a subsequent section.
Pathologists see the range of changes described by more systematic research of the trophoblast remodeling of the spiral arteries. The infiltration of the decidua by giant trophoblast cells is often a prominent feature. Plugging of spiral arteries can be seen. Endovascular invasion of spiral arteries may be present (Figs 3,4).
Fig 3: The top portion of decidua demonstrates embedded endometrial glands with plugged mucus in the lumen. Beneath them are two profiles of a remodeled spiral artery. On the left, the arrows point to cytotrophoblast replacing the muscularis. To the right the endothelial lining has been replaced by cytotrophoblast (CT). There are cytotrophoblast, including multinucleated cells surrounding the vessels. At the bottom is the basal cytotrophoblast/fibrinoid layer, and the immature villi can be seen below that band (V). H&E, 10x
Fig 4: The arrows show the brown immuno-staining that identifies endothelium lining a vascular space, presumably a spiral artery devoid of muscularis. The large irregular cells in the lumen are typical cytotrophblast within the vessel that are undermining the endothelium, and will replace it. CD31
The findings are a mix of necrotic decidua capsularis, which occurs over the implantation, the decidual basalis that is beneath the placenta and the parietal decidua beneath the membranes. Elective abortion curettage has not been available to me for examination, so there is no normal gestation matched control tissue. There is decidua from medical induced abortions for lethal anomalies, but I have not systematically compared them to examples with known causes of fetal loss. I attempted a comparison of fetal loss curettage findings in diabetic pregnancies with non-diabetic, and did not find significant differences7.
The uterus may be surgically removed following delivery for uncontrollable uterine bleeding, usually due to some form of placenta accreta, but occasionally for uterine atony. In the latter, the uterus fails to contract despite therapeutic injection of uterine stimulants and physical massage. Less commonly, the uterus is included in the autopsy of a maternal death. Forensic pathologists may also examine gravid uteruses from causes of death not directly related to obstetrical disease such a trauma or homicide. There have been some publications of systematic reviews of whole gravid uteri 8and of maternal autopsies with toxemia 9.
I regret not having explored and recorded more of the vascular anatomy in the uteri that I examined. The focus was on the diagnostic features of accreta, or possible causes of atony. One incidental observation that I could not explain in a uterus removed for placenta accreta was severe chronic inflammation in a deep myometrial artery in which a portion of the artery was necrotic suggesting trophoblastic remodeling (Figs 5a, 5b).
Fig 5a: The artery in the center is surrounded by large gravid myometrial myocytes. Its wall on the right is intact, but the left, top and bottom shows loss of medial smooth muscle, numerous inflammatory cells, and smudged eosinophilic material in the media. H&E 10x
Fig 5b: This image is a higher magnification of the above artery at the junction of intact and damaged media. The inflammatory cells appear to include neutrophils, lymphocytes and even plasma cells. H&E 40x
Un-remodeled muscular spiral arteries are usually found in the membranes. They may be difficult to distinguish from the deeper, smaller radial arteries. The most striking pathologic lesion of these spiral arteries is acute atherosis, which will be discussed below. The arteries may demonstrate numerous mononuclear cells in or around the wall of the spiral arteries, or even eosinophils (Figs 6a, 6b, 7).
Fig 6a: This is a low power image of a spiral artery cut in the parallel plane of the delivered parietal decidua. Various profiles of a thin walled spiral artery can be seen, and many are surrounded by inflammatory cells. H&E 10x
Fig 9: This is another example of a thrombus in the parietal decidua showing a thrombus with medial necrosis of the vessel, likely a spiral artery. There is very early organization of the thrombus. H&E 20x
The spiral arteries of the membranes even in term pregnancies may show changes of trophoblastic remodeling of spiral arteries. The chorionic epithelium of the membrane may appear to have loosened and descended into the trophoblast (Figs 10a, 10b, 11,12).
Fig 10a: This sample of parietal membrane demonstrates loss of the intact cytotrophoblast layer with many independent often multinucleated trophoblast cells (arrows). There are multiple profiles of a spiral artery. Some are normal, and others show dilatation, thinning and necrosis of the media and focal inflammation (*). H&E 10x
Fig 10b: This higher power of the area of the * in the previous figure shows the eosinophilic necrosis of the media, and the mononuclear inflammation. The red cells appear distorted suggesting stasis. H&E 40x
Fig 11: This spiral artery in the parietal decidua demonstrates destruction of the media partially showing early organization (scar formation) and a lining of cytotrophoblastic cells (dotted arrow). Loss cytotrophoblast are present in the decidua (arrow). H&E, 20x
The spiral arteries may have a thickened media, which has been considered one form of decidual vasculopathy. I do not know of specific, quantified diagnostic criteria correlated with clinical income. Spiral arteries may appear thickened because of cell swelling, interstitial edema and interstitial fibrinoid material, as well as by smooth muscle hyperplasia10. The normal thickness may depend on gestation as well as nearness to areas of invasive cytotrophoblast (Figs 14-18).
Fig 14: Low power of a spiral artery in the parallel plane of the parietal decidua showing a media of sparse myocytes of which the thickness appears to vary because of oblique cutting of the vessel. The mother suffered from diabetes and chronic hypertension. H&E 4x
Fig 18: This higher power of the above sample shows that the apparent thickened spiral artery is due to swelling from trophoblast-induced damage (circle). The trophoblast can be seen in the margin (arrow) H&E 10x
The spiral arteries at the base of the placenta have been transformed by trophoblast and often are only multiple profiles of a tortuous vessel shell lined by a thick hylaine material, usually with embedded cytotrophoblast cells. As in the decidua of the membranes, there may be basal arteries. Untransformed spiral arteries may be present and have been considered a possible marker for under-remodeling of spiral arteries.
A perhaps understudied aspect of the spiral arteries in the basal decidua is their appearance when they underlie lesions of placental infarction, retroplacental hematoma, subchorionic thrombohematoma, and infarction hematoma. Those underlying infarctions may be distended with blood and thrombus (Figs 19-24).
Fig 19: The large, blood filled convoluted spiral artery at the base of the placenta shows lamina of fibrin. The infant was liveborn at 26 weeks of gestation, but was severely growth restricted. The mother had type 1 diabetes and chronic hypertension. The placenta had multiple infarctions. (H&E, 2x)
Fig 20: This view of the same spiral artery shows is proximity to a recent placental infarction. The presence of a thrombus in the vessel suggests that the infarction may have been caused by this very superficial thrombus, but there is no obvious clue as to how it formed. (H&E, 2x)
Fig 22: In another area of the above vessel the infarction involves the basal decidua (necrosis) as well the villi. Surrounding viable appearing decidua shows acute hemorrhage between the decidual cells. A gross retroplacental hematoma was not present. (H&E, 10x)
Fig 23: The base of the placenta demonstrates a dilated spiral artery with laminated fibrin and an acute overlying placental infarction. Around the spiral artery hemorrhages can be seen in the basal decidua. The placenta had multiple infarctions and this small retroplacental hematoma. The infant was small for gestation with Apgar scores of 7 &8. (H&E, 2x)
Fig 24: A high power over the base of the placenta above the retroplacental hematoma demonstrating an influx of neutrophils, a typical inflammatory response to a retroplacental hematoma (arrows). (H&E, 40x)
Fig 25: The infarction shows the compacted pale villi in the lower portion of the image. Above the infarction a complex spiral artery profile shows lumen empty except for some faded blood cells. (H&E, 2x)
Fig 26: This is another infarction, oriented above the base of the placenta, which runs across the lower left corner of the image. In the lower left of this diagonal can be seen a dilated spiral artery that appears empty except for faded blood cells. There is no thrombus at this level of the superficial decidua, although the infarction is evidence of vascular occlusion in the spiral artery. (H&E, 2x)
Fig 28: This is similar to the above two cases. The infarction to the left with the base of the placenta shows above it in the image. The base demonstrates a long, complex set of cross sections of a spiral artery showing old, pale red cells. (H&E, 2x)
The implication is that the occlusion of the vessel, mostly likely from thrombus, occurred proximal to the portion of spiral artery in the most superficial dedicua. In retroplacental hematoma, vessels may demonstrate rupture and hemorrhage into the underlying decidua. In infarction hematoma, a spiral artery may appear to rupture into the intervillous space. In most cases of these complications, no spiral artery is identified, likely because it has not been sampled, or is no longer attached to the placenta.
The lining of base of the placenta on the intervillous surface appears to have an endothelial phenotype, as do vessels that are likely veins. The arteries maintain a trophoblast phenotype, but more systematic studies may show less of a dichotomy (Figs 30,31).
Fig 30: This normal mature placenta demonstrates villi with brown staining from an antibody to placental alkaline phosphatase (PLAP) present on the surface of the villous trophoblast. The basal decidua runs along the right of the image. The spiral arteries are also lined by brown stained cells indicating that the lining cells are trophoblastic. (PLAP, 20x)
Fig 31: This is the same placenta showing brown immunostaining with an antibody to platelet endothelial cell adhesion molecule (CD31) that stains endothelial cells in the fetal vessels in the villi, and the lining of a vessel in the decidua, which may be a large vein, but a deeper portion of a spiral artery cannot be excluded. (CD31, 10x)
The goal of the placental pathologist is to find lesions that either explain the outcome of the current pregnancy, or even better predict the problems of the next pregnancy or of possible subsequent maternal disease. Subsequent sections will discuss toxemia (and the varied terms for this disease(s)) and discuss the causes of fetal growth restriction. However, both of these clinical complications may be related to a failure of trophoblast remodeling of the spiral arteries. Hence, logically, pathological examination of these arteries may clarify mechanisms of these complications.
The origin of the failed spiral artery remodeling is still unresolved. In a 2010 review article that presents numerous control factors that may be involved in trophoblast remodeling, the authors lament that the mechanisms that these interdependent mechanisms have primarily been considered in isolation11. They conclude It is only when we have a clearer picture of the events occurring in a normal pregnancy that we can begin to determine which of these are compromised in pregnancies complicated by disorders such as pre-eclampsia. In the subsequent almost one decade, there have been studies elucidating the roles of natural killer cells, local hemodynamics, and numerous biological signals including the effects of certain microRNAs on tissue culture models. I still cannot see a comprehensive quantitative model of the process of trophoblast remodeling. I think achieving that model would be a breakthrough that would have implications for more that preeclampsia. Yet, to understand the origin of preeclampsia, at least to me, looking for upstream events that differentiate first pregnancies from subsequent pregnancies might provide some therapeutic leverage even if the understanding of trophoblast modeling is incomplete.
The most clinically salient clue as to the origins of toxemia is that it predominates in the mothers first pregnancy. This suggests that looking at key differences between the first and subsequent pregnancies might point to the initiating pathogenesis. One clue that simple early hormonal differences could play a role is provided by primate studies that administer first trimester estrogen to interfere with trophoblast remodeling of spiral arteries12. A delay in starting the remodeling might be enough to subvert it. I do not know if first pregnancies are slower to elevate progesterone over estrogen, but hormonal differences could be important. Another upstream observation is a reported decrease in hyperglycoslated HCG in toxemia 13. The suggestion is that hyperglycoslated HCG prevents TGF beta induced cytotrophoblast apoptosis and promotes invasion. Finding the upstream events initiating toxemia is a reasonable approach until we can comprehend the regulation of the whole process.
This thought allows re-phasing the original question as can the examination of the spiral artery demonstrate early mechanisms of failed trophoblast remodeling. One potential study is to compare the spiral artery findings in first trimester primigravida pregnancies to those in multiparous mothers, and look for statistical differences. Most of the specimens that I have seen are from spontaneous abortions that may not represent normal trophoblast remodeling, but this approach could be productive if such specimens were available. Another approach is to examine delivered placentas to find evidence of difference between primigravida and multigravida placentas, or between patients with different clinical patterns of toxemia. Stanek reported observations of the placenta and membranes after delivery to identify shallow implantation which I interpret to be the same as the incomplete trophoblast remodeling14. These features included those of spiral arteries including hypertrophic arteriopathy and atherosis, as well maternal floor trophoblastic giant cells and excessive numbers of extravillous trophoblast. These features distinguished normal from preeclamptic (toxemic) pregnancies, but not early from late onset disease. The implication is that these clinically different patterns do not differ histologically on key features of the spiral artery changes. This study has large numbers of cases recorded by an experienced pathologist but the conclusions would be strengthened if tested in a blind prospective study with concordance of diagnoses among a group of pathologists.
Another effect of toxemia is an increased prevalence of placental infarctions and retroplacental hematomas including those producing clinical abruption of the placenta. While these lesions will be considered in other sections and chapters, looking at the spiral arteries underlying these lesions may be of value. The underlying causes of infarction and abruption in toxemia may be the same or different from that occurring in other clinical disease. These differences could have predictive value for future pregnancies. I am unaware of a successful attempt to make these distinctions, but there is more work to be done.
The one distinct anatomic diagnosis of the spiral arteries in pregnancy is acute atherosis. Its clinical significance is still uncertain. This lesion is part of a triad that is accepted as decidual vasculopathy15. Two of the components I will lump together as acute atherosis, the shiny eosinophilic fibrinoid necrosis of the arterial wall and the accumulation of lipid filled macrophages in the fibrinoid (Fig 32,33).
Fig 32: The specimen is from the decidua attached to the fetal membranes in a mother with toxemia. There are three cross sections through a spiral artery. The cross section on the right shows a dilated portion with loss of muscular wall. Where the muscular wall would have been there is a thick, pink (eosinophilic), smudged appearance. This is the fibrinoid necrosis. Within this structure, the arrows point to foamy clear macrophages. There are also scattered mononuclear inflammatory cells.
Acute atherosis is a lesion of untransformed spiral arteries. Fibrinoid is a pathologic term that simply implies that under the microscope, the pink amorphous material looks like fibrin that has deposited, but it does not imply actual thrombus formation. There is still some debate about the nature of the pink material. The necrosis of the smooth muscle, increased endothelial permeability to plasma proteins and complement deposition has all been invoked as contributors to the appearance. The lipid cells have been identified as macrophages based on CD68, a lysosomal marker. The bulk of the lipid presumably comes from transudation of serum proteins, analogously to atherosclerosis. The role of inflammatory cells, which may be in or around the vessel wall, is unknown. Often there are no inflammatory cells in direct proximity to the lesion. The lesion is often segmental involving only some histologic profiles of the artery. The vessel is usually dilated.
As far as can be discovered, including in Sheehan and Lynchs monograph of 677 maternal autopsies of mothers with toxemia, acute atherosis is restricted to uterine spiral arteries9. (Sheehan and Lynch did find infrequent infiltration of arterioles with diffuse lipid with fat stains in the kidney and liver, but without vascular necrosis.) Any theory of the pathogenesis of acute atherosis would have to explain not only the anatomic restriction to spiral arteries but other observations. First, why is the lesion is segmental in one artery, but also why does it only involve some of the arteries in the same decidual area. Second, why do only a minority of women with toxemia demonstrate the lesion? Third, why is the lesion present in some patients without clinical toxemia. Spiral arteries are unique not just because of acute atherosis but also that during the menstrual cycle they respond to hormonal stimuli, and finally shed to regrow with the next cycle. The decidual environment is also unique and of perhaps relevance, decidual cells appear to produce renin, the initiating enzyme in the vasoconstrictive angiotensin system.
An important attempt to explain acute atherosis was published in a review article by Carlos Labarrere who was working with Paige Faulk at the Center for Reproduction and Transplantation Immunology at Methodist Hospital in Indianapolis IN, when I met him16. He reviews the work of the English group on placental bed biopsies and the hypothesis that PIH, as well as unexplained IUGR, is associated with placental arteries that have not undergone trophoblast invasion. He also considers the studies of acute atherosis observed in diabetes, hypertension without PIH, unexplained IUGR and autoimmune disease. He points out fluorescent studies of atherosis show large IgM and C3, and some C1q. He says similar deposits are seen in rejection lesions with lipid deposition. In describing rejection, he points out lipid starts in myointimal cells and is part of insudation from damaged endothelium. The finding of IGM and complement in acute atherosis is consistent with an immune complex disorder perhaps related to a fetal antigen. The insudation of lipid does not appear to be simply leakage of serum proteins through the damaged endothelium. Immunofluorescent studies of acute atherosis did not localize albumin or transferrin in the vessels. Dr. Labarrere argues that with an abnormal immune response, fetal antigens in excess would generate IgM instead of a protective IgG response. He cites some serum evidence of immune complex disease in PIH that I have not reviewed. Dr. Labarreres review was published in 1988, yet many of the same questions remain today.
A note about his cited immuno-fluorescent study published in 198117. The incidence of acute atherosis in patients with preeclampsia was 53%, which is high. They also found the lesion in patients with normotensive insulin dependent diabetes and with stable chronic hypertension. This study was done on curettage specimens based on palpating the location of the placenta before delivery, and then scraping that area after delivery. The diagnosis of acute atherosis was made on frozen samples. They recognized that trophoblast transformed vessels could also show lipid in from degenerated cytotrophoblast in the walls of these vessels. Controls opposite the placental bed and in normal pregnancies did not show immunostaining. Since insulin dependent diabetes could be associated with endothelial damage, the lesions resemblance to that seen in preeclampsia is not surprising. I have not seen the lesion in fetal membranes in this context. However, more aggressive management of serum glucose in these patients may have decreased the underlying vascular pathology. The cases of acute atherosis in some normotensive patients with systemic lupus erythematosis would support an immune complex derivation of the atherosis. I wonder if redoing of immuno-localization studies using antibodies to C4d might not help clarify whether immune complexes are involved in the pathogenesis. Using arteries from the fetal membranes would reduce possible diagnostic confounding with trophoblast remodeled vessels.
Acute atherosis occurs in only a minority of patients with toxemia. A potential explanation for this observation is that the prevalence of a focal lesion is being underestimated as a result of limited sampling. Most of the untransformed spiral arteries seen by the pathologist are in the decidua attached to the fetal membranes. Pathology assistant Yasser Daoud and I presented a small study at the first Mid America Placenta Study Group meeting in Toronto in 1998 that tried to answer this problem. We sampled the fetal membranes by taking a 1-2 cm roll of membrane from the rupture site to the placental margin. Two cross sections of a single membrane roll were included in a histology-processing cassette. In 25 cases with a clinical diagnosis of pregnancy-induced hypertension, three additional membrane rolls were processed. Nineteen patients had no atherosis on any of 4 slides, 3 had atherosis on one slide, and three had atherosis on two slides. In two cases, the first sample had the atherosis. Further sections diagnosed four additional lesions. (Interestingly, six placentas had one or more infarctions, of which three had acute atherosis.) Our study demonstrated that increasing the sample size did increase the number of positive cases, as would be expected in a focal lesion. Since three cases had two positive slides out of 4 samples, and 19 cases had no positive slides out of 4 samples, our results suggest a real difference in the density of the lesion among patients. We could not conclusively answer if there were cases that truly had no atherosis. We decided to try a different approach of identifying acute atherosis grossly, which will be described below.
The incidence of decidual vasculopathy varies among investigators, in part because not all studies may include all three potential components, fibrinoid necrosis of the vessels, lipid laden cells in the necrotic material, and separately muscular hypertrophy. I have not discussed the latter because while tempted at times to diagnose the lesion, I have hesitated because quantitative lesions are difficult to judge, because my subjective experience is that the thickness is more in younger gestation placentas, and because I not sure that I am not looking at radial arteries or a transient change in a vessel. However, most studies have concentrated on the two lesions that I am lumping for this web page as acute atherosis. The rational for an overall designation of decidual vasculopathy can be found in the placenta consensus conference monograph15.
Recently I received an email from Dr. Peilin Zhang with some publications he had written about acute atherosis18(http://dx.doi.org/10.1101/293027.). He demonstrated cytokeratin (AE1/AE3) and , CD56 staining in the foam cells in acute atherosis. He argued that the assumption that the lipid cells in acute atherosis were macrophages was based on morphologic analogy with atheromas in arteriosclerosis and by CD68 staining, and is wrong. I have to admit that CD68 is not specific for macrophages, and in fact I had often seen that is was immuno-positive in cytotrophoblast cells (Fig 34).
Fig 34: This image of placenta demonstrates massive chronic intervillositis with the many monocytic cells stained deeply dark brown for the antigen CD68. The arrow demonstrates the paler, but still positive for CD68 cytotrophoblastic cells embedded in the surrounding fibrinoid. (CD68, 20x)
I have never tried to identify the cells in acute atherosis with cytokeratins or CD56. However, as I suspected from his publication photograph, and even more clearly in his preprint, the vessels he illustrates are surrounded by cytotrophoblast. I would have considered them partially remodeled vessels in the basal decidua, and not acute atherosis. His prevalence of the acute atherosis/fibrinoid necrosis is high (63%) and from the published illustrations, he evaluated basal decidua.
The consensus monograph of placental diagnosis allows inclusion of unremodeledspiral arteries in the base of the placenta. I only make the diagnosis of acute atherosis in vessels at the base of the placenta if they clearly are not in proximity of cytotrophoblast. In almost all cases that I have diagnosed as acute atherosis, I saw the lesion in the fetal membranes. Given the wide range of reported prevalence of acute atherosis, I suspect that there is poor concordance among pathologists for the diagnosis of acute atherosis in the base of the placenta. It would be interesting to know if the pink hyaline is different in remodeled arteries, which apparently includes fibrin, degenerated trophoblast, and likely secreted cytotrophoblast matrix, from that in acute atherosis 10. A reliable immuno-histological stain to distinguish the two processes would be diagnostically useful.
The importance of the distinction between acute atherosis and trophoblast remodeled vessels is more than academic. This concern is evidenced in a study that compared hemodynamic measurements in preeclamptic mothers seven months after delivery with and without decidual vasculopathy19. The study definition of decidual vasculopathy was vascular fibrinoid necrosis and lipid filled foam cells in the vascular wall of spiral arteries in either the decidua basalis or parietalis. The basalis refers to the arteries under the placenta, and parietalis to those in the fetal membranes excluding placenta. This inclusion of both decidual regions likely accounts for the high incidence of 53.4% decidual vasculopathy in their patients and likely includes some specimens with only basilar trophoblastic remodeling of arteries. Importantly, using this definition they found significant differences in some cardiovascular parameters between those patients with and without decidual vasculopathy. The long-term significance of decidual vasculopathy to the mothers health could not be determined from this study, but the authors demonstrated an increase in hypertension in these patients, and in an earlier publication correlated decidual vasculopathy with some measures of more severe obstetrical disease, including higher diastolic blood pressure, lower birth weight, and lower umbilical cord pH20-22.Their findings may be partially confounded by bundling two distinct histologic entities.
What is the correct definition of decidual vasculopathy? These clinical papers and Dr. Zhangs paper led me to question the criteria that I use for the diagnosis of acute atherosis, or as Dr. Zhang terms it classic decidual vasculopathy (This excludes muscular hypertrophy as part of the diagnosis). I think a consensus study of shared slides and multiple placental pathologists would better define the nosology of the lesion. I think we may need either some new definitions or a broader classification. I am particularly interested in the relation of lesions seen in the basal placenta to those in the membranes. Are the lesions in the base of the placenta the best indicator of poor spiral artery remodeling? Is acute atherosis seen in the free fetal membranes not a unique lesion, but a failed trophoblast modification of a spiral artery far from the implantation?
The last question was suggested to me in a preliminary, unpublished study of spiral arteries in decidual membrane that I did again with Yasser Daoud. The study attempted to improve our ability to sample acute atherosis in the fetal membranes. The entire membranes after removal from the placental margins were stretched over white cardboard cards, and a dissecting microscope was used to reveal the irregular spiral arteries. We sampled the visible spiral arteries for histological examination and made note of dilated segments that were suspected of being acute atherosis (35).
Fig 37: The vessel identified as number 10 demonstrates dilated segments of spiral artery from medial muscle destruction, but this is due to cytotrophoblast that has migrated into the parietal decidua (arrows). (H&E, 10x)
If anyone wants to complete such a study, one measure to record is the distance of the sampled vessel from the placenta. This could be an important parameter determining the prevalence of decidual vasculopathy.
The explanation of the absence of acute atherosis in some patients with toxemia could also be that since acute atherosis is not directly involved in the pathogenesis of toxemia, different patients may have inherent differences in the susceptibility of their spiral arteries to the lesion. Another simple alternative explanation is that those with atherosis may have a different response to the incomplete spiral artery remodeling with perhaps of a different mix of anti-epithelial signals. The pathogenesis of toxemia will be discussed in a later section.
We dont know what causes acute atherosis in toxemia, but the pathogenesis may not be unique to that disease. A different disease might produce the lesion by the same pathologic mechanism or it might through a different pathogenesis produce an appearance that mimics the pathology seen in toxemia. While I was still a fellow in pediatric pathology I examined a placenta from a mother with a clinical diagnosis of TTP that had pink focal lesions of the inner surface of spiral arteries that we suspected were due to small platelet clots on the endothelium causing necrosis of the adjacent media. We sent the slides to a respected consultant who said we were just seeing acute atherosis. I still have my doubts but I cannot find images from that case.Others have since reported TTP as an acute atherosis mimic23. In other cases, I have seen necrotic vascular lesions that are of unknown etiology even in normal pregnancies that I have diagnosed as fibrinoid necrosis of spiral arteries for lack of a better term. These lesions could have a different etiology from those in toxemia. However the resemblance still needs an explanation (Figs 38,39).
Fig 38: This low power image of the membranous decidua sectioned tangential to the membrane plane shows in the convoluted cross sections of a spiral artery, which appears dilated in the right corner of the field. The placenta was uncomplicated with a 3000 kg infant at term with no maternal history of toxemia or autoimmune disease. (H&E, 4X)
In acute atherosis with fetal growth restriction, the question I think becomes, at least in infants whose placenta shows evidence of utero-placental ischemia, why they do not have clinical toxemia. In diabetic mothers with acute atherosis a mimic lesion from endothelial injury from glucose arises as a possibility. In immune complex disease such as SLE the mechanism is again likely to be endothelial injury with an unexplained predilection fro spiral arteries. Studying in detail the differences in acute atherosis with different diseases potentially could yield new insights into the pathogenesis of the lesion.
In the papers that Dr. Zhang sent to me, he argues that CD56 is key to acute atherosis, and that the antigen is in some way transferred from CD56 positive natural killer cells. There certainly is a literature on the role of natural killer cells in pregnancy, and Dr. Zhang reviews some interesting studies in mice. I recall a paper presented at a symposium at the University of Kentucky on comparative studies of placenta on the importance of natural killer cells in pig pregnancies. In the Guinea pig, the mesometrial arteries are remodeled by non-trophoblastic mononuclear cells 24. In vitro research points to matrix degrading proteases produced by natural killer cells that migrate to spiral arteries in the decidua 25. However, I currently cannot accept the CD56 vasculopathy hypothesis, but it made me consider two things. The first is that anti-CD56 marks cytotrophoblast is a not a surprising finding, given that this is a common surface antigen of embryonic tissue and has been known since at least 1994 26. However, it may be important that CD56 is a cell adhesion molecule (once thought to neuro-specific, hence the alternative name, N-CAM), and that it can bind to itself. The first consideration is can natural killer cells bind to cytotrophoblast via CD56?
The second consideration is the potential value of understanding the evolutionary play between invasive trophoblast cells, which mimic malignancy in many ways, and the role of natural killer cells. I agreed with Kurt Benirschke that we needed to better understand the evolutionary biology of the placenta from pre-placenta forms for example guppies to the wide variation in placentation in mammals. He was way ahead of me, and years before his recent death, he started an important web site of comparative placentation. The evolution of natural killer cells in the endometrium might shed light on their role in human pregnancy and malignancy.
Toxemia is often separated into an early onset versus a late onset usually milder disease. The major complications in the mother are eclampsia (seizures), liver and cerebral hemorrhage, and placental abruption. The fetal complications are growth restriction and its potential complications including stillbirth. Some studies have found an increased incidence of complications with acute atherosis.
Pathologists see the residua of previously remodeled spiral and radial arteries in uteri removed for other causes later in a gravid womans life. (Spiral arteries in the superficial decidua are shed at the end of delivery.) How is this process of involution controlled? Is it simply the removal of progesterone or other signals that were associated with the placenta? How are new radial and deep spiral arteries regenerated? As per Dr. Zhang, is there a role of natural killer cells in removing intravascular cytotrophoblast, since both they and the cytotrophoblast share an adherence molecule, CD56?
These questions bear on the question as to why primigravidas are at the highest risk of toxemia. As a resident, I recall speculation that it must be due to acquired immuno-tolerance. Some papers even suggested that the risk increased with a different father of the fetus. However, that notion was dampened by a Norwegian epidemiologic study published in the New England Journal in 200227which demonstrated the interval between pregnancies was the determining factor. A related phenomenon is that the firstborn child usually has the lower birth weight of his/her siblings. To most observers, this suggests that even without diagnostic toxemia, the firstborn child likely is less successful establishing a utero-placental circulation than his later siblings. Dr. Yee Khong who made important contributions to the concept of inadequate remodeling of spiral arteries to the development of toxemia, tried to understand this primigravida phenomenon by a direct observation. He and colleagues using a random sample from the superficial myometrium of non-gravid hysterectomy specimens, compared the extent of smooth loss in the media, and of elastic tissue fragmentation of spiral arteries in women of different gravida28. They found significant evidence of elastica degeneration and loss of medial muscle between nulliparous women and those who had children. As noted in the paper, they could not determine if the sampled arteries had been beneath the placenta. The assumption was that within sampling error, the effect was still evident because the arteries had been modified in the past by trophoblast remodeling within the placental bed. If prior pregnancy weakens spiral arteries and accounts for larger fetal growth and less toxemia in subsequent pregnancies, is it possible that the extent of deep trophoblast remodeling of spiral arteries extends beyond the immediate placental bed. It would be interesting to plot not only the extent of finding cytotrophoblast beyond the placental margin, but also the pattern of vascular remodeling in uteri removed after pregnancy for disease unrelated to pregnancy.
The spiral artery remodeling is a very complex process but critical to lowering vascular resistance and allowing the necessary increase in uterine blood flow to the placenta. Toxemia is likely due to a primary failure to obtain sufficient high flow through the intervillous space. The basis for this failure is still moot, but the process may be initiated early in pregnancy. One common consequence is the placental and fetal responses to decreased utero-placental blood flow, which will be discussed in Section 9c and 9E. The placenta in toxemia induces active signals in the maternal blood that oppose endothelial growth factor and results in damage endothelial cells. The lesions in the mother will be discussed in Section 9b.
There are many unanswered questions relating the spiral artery lesions to toxemia. Why is toxemia more common in the first pregnancy and in molar and diandric partial (triploid) molar pregnancies? What is the relationship of acute atherosis to spiral artery remodeling and toxemia? How does the failure of trophoblast remodeling of spiral arteries predispose to placental infarction and retroplacental hematoma?
For the diagnostic pathologist, the major question is whether any lesion, acute atherosis or otherwise, has any prognostic value for future pregnancies or future cardiovascular disease in the mother. Prospective long-term clinical pathological correlation studies are needed to answer this question. These studies need to utilize accepted pathological criteria. Further basic molecular and experimental studies of trophoblastic remodeling of vessels is likely to provide valuable insights not only into toxemia, and utero-placental ischemia, but also into cellular pathways of trophoblastic invasion that could be coopted in malignancy.
Uterine spiral arteries play a vital role in supplying nutrients to the placenta and fetus, and for this purpose they are remodelled into highly dilated vessels by the action of invading trophoblast (physiological change). Knowledge of the mechanisms of these changes is relevant for a better understanding of pre-eclampsia and other pregnancy complications which show incomplete spiral artery remodelling. Controversies still abound concerning different steps in these physiological changes, and several of these disagreements are highlighted in this review, thereby suggesting directions for further research. First, a better definition of the degree of decidua- versus trophoblast-associated remodelling may help to devise a more adequate terminology. Other contestable issues are the vascular plugging and its relation with oxygen, trophoblast invasion from the outside or the inside of the vessels (intravasation versus extravasation), the impact of haemodynamics on endovascular migration, the replacement of arterial components by trophoblast, maternal tissue repair mechanisms and the role of uterine natural killer (NK) cells. Several of these features may be disturbed in complicated pregnancies, including the early decidua-associated vascular remodelling, vascular plugging and haemodynamics. The hyperinflammatory condition of pre-eclampsia may be responsible for vasculopathies such as acute atherosis, although the overall impact of such lesions on placental function is far from clear. Several features of the human placental bed are mirrored by processes in other species with haemochorial placentation, and studying such models may help to illuminate poorly understood aspects of human placentation.
Deep placentation in the human requires physiologic transformation of the spiral arteries into uteroplacental vessels. This process involves the inner myometrial segment (junctional zone) of the spiral arteries and is effected by trophoblast invasion of the vessel wall, resulting in complete loss of the arterial structure and deposition of fibrinoid and fibrous tissues. Absent or inadequate physiologic changes in the junctional zone spiral arteries limits placental blood flow in pregnancies complicated by preeclampsia and fetal growth restriction. The cause of defective deep placentation is still unknown, although it is often attributed to impaired trophoblast function and migration. However, trophoblast invasion is preceded by decidual remodeling of maternal tissues, a process that is initiated in the endometrium but extends into the junctional zone. This review examines the mechanisms that control decidualization and subsequent trophoblast invasion in normal and abnormal pregnancies. The possibility that disruption of the decidual process in the secretory phase of the menstrual cycle triggers a cascade of events resulting in failed deep placentation is explored. (Am J Obstet Gynecol 2002;187:1416-23.)
Reprint requests: Jan Brosens, MD, Institute of Reproductive and Developmental Biology, Imperial College School of Medicine, Hammersmith Hospital, London W12 0NN, United Kingdom. E-mail: [emailprotected]
Extravillous trophoblast remodeling of the uterine spiral arteries is essential for promoting blood flow to the placenta and fetal development, but little is known about the regulation of this process. A defect in spiral artery remodeling underpins adverse conditions of human pregnancy, notably early-onset preeclampsia and fetal growth restriction, which result in maternal and fetal morbidity and mortality. Many in vitro studies have been conducted to determine the ability of growth and other factors to stimulate trophoblast cells to migrate across a synthetic membrane. Clinical studies have investigated whether the maternal levels of various factors are altered during abnormal human pregnancy. Animal models have been established to assess the ability of various factors to recapitulate the pathophysiological symptoms of preeclampsia. This review analyzes the results of the in vitro, clinical, and animal studies and describes a nonhuman primate experimental paradigm of defective uterine artery remodeling to study the regulation of vessel remodeling.
During early human pregnancy, extravillous trophoblast (EVT) migrates to, invades, and replaces the vascular smooth muscle cells (VSMC), endothelial cells, and elastic lamina within, thereby remodeling the uterine spiral arteries [1,2,3,4]. Consequently, these arteries change from high-resistance/low-capacity to low-resistance/high-capacity vessels, and thus, uterine artery blood flow increases with advancing gestation to enhance placental perfusion and promote fetal development. Defective uterine artery remodeling (UAR) underpins the etiology of certain pregnancy disorders that comprise the syndrome of placental ischemia [5, 6], notably early-onset preeclampsia, defined as premature delivery prior to week 34 of gestation and associated with a high rate of fetal growth restriction [7,8,9,10,11,12,13]. The term preeclampsia is used throughout this review to refer to early onset since in contrast to late-onset preeclampsia, i.e., delivery after 34weeks, it is underpinned by defective UAR. Preeclampsia is associated with maternal systemic vascular endothelial inflammation-activation-dysfunction, hypertension, renal glomerular endotheliosis, and proteinuria, as well as maternal and neonatal morbidity/mortality [14,15,16,17,18,19,20]. It has been proposed that as a consequence of impaired UAR and placental perfusion, the placenta exhibits oxidative stress and the release of anti-angiogenic factors, cytokines, and/or syncytial extracellular vesicles which, along with predisposing maternal factors such as obesity and hypertension, elicit the pathophysiological manifestations of preeclampsia (reviewed in [16, 20]). Excessive trophoblast invasion and UAR are also deleterious because they result in impaired uterine artery vasomotor tone and hemorrhaging after delivery, a pregnancy complication known as placenta accreta [21, 22]. Despite the fundamental importance of UAR to successful pregnancy and fetal development, relatively little is known about the regulation of this process primarily because the majority of studies have focused on the pathophysiological consequences of adverse conditions of pregnancy and not on UAR. The present review describes the results of the in vitro and in vivo studies and a nonhuman primate model to study the regulation of UAR.
Numerous in vitro studies have been conducted to investigate the ability of primary or immortalized trophoblasts cultured in two or three dimensions to pass across a synthetic permeable membrane coated with matrigel or decellularized extracellular matrix or to form endothelial-like tubes as indices of cell migration and invasion. Collectively, these studies have shown that several factors known to be produced by the placenta and/or decidua, including vascular endothelial growth factor-A (VEGF), placental growth factor (PlGF), insulin-like growth factor (IGF), epidermal growth factor (EGF), heparin-binding EGF (HB-EGF), activin, and human chorionic gonadotrophin (hCG), stimulated HTR-8/SV neo, trophoblast, or choriocarcinoma JEG-3 cell migration or endothelial-like tube formation [23,24,25,26,27,28,29,30,31,32,33]. Moreover, transcription and cell signaling molecules, including the Rac1 member of the Rho family of GTPases, the elastin-derived matrikine VGVAPG, the ephrin-B2 ligand of the Eph receptor, and Notch-2, also increased trophoblast migratory capacity in vitro [34,35,36,37,38]. However, in other in vitro studies, several of these factors did not alter EVT migration [39,40,41]. In contrast, transforming growth factor (TGF)-1, TGF-2, and TGF-3 and endocrine gland VEGF (EG-VEGF), as well as microRNA-93 and microRNA-135 which decrease CXCL12 gene expression, inhibited migration/invasive capacity of trophoblasts [42,43,44,45,46,47], while inhibition of TGF3 restored invasive capacity of trophoblasts obtained in late gestation from placentas of women with preeclampsia . Additional in vitro studies using placental explants showed that elastin-derived peptides increased and endothelin-1 decreased trophoblast overgrowth [35, 49]. The underlying causes of the divergent effects of these factors on trophoblast migration are unclear, although use of different culture conditions, including oxygen and hypoxia-inducible factor levels and transformed versus primary trophoblasts, may be involved.
The presence of uterine natural killer (uNK) cells and macrophages, which are sources of VEGF-A and VEGF-C, angiopoietins, interleukins, and matrix metalloproteinases (MMPs) , was associated with VSMC and endothelial cell disruption in decidual tissue obtained in early human pregnancy , while the addition of uNK cell-conditioned medium to cultures of human term chorionic plate arteries caused VSMC and extracellular matrix breakdown . The addition of EVT-conditioned medium to cultures of vascular endothelial cells increased expression of the chemokines CCL14 and CXCL6, which induced chemotaxis of decidual NK cells and macrophages, and the authors proposed that there was crosstalk between EVT, endothelial cells, and decidual immune cells in spiral artery remodeling . NK cells also enhanced migration of and tube formation by primary trophoblast cells from placental villous tips, an effect that was prevented in cultures containing NK cells pretreated with sphingosine FTY720 to suppress NK cell function and VEGF production . Moreover, recent studies suggest that additional processes, including invasion of uterine veins and lymph vessels by endo-venous and endo-lymphatic trophoblast cells, respectively, may also be involved in uterine artery remodeling [55, 56]. Based on these studies, it has been proposed that the immune system plays a role in uterine vessel transformation, although it has been suggested that the role of the immune system is more established in mouse than in human pregnancy (reviewed in [57,58,59]).
Clearly, the in vitro studies are significant in showing that a multitude of factors have the capacity to alter migratory and invasive capacity of trophoblast cells. However, considering the highly complex interplay of different cell types, molecular events, and spatio-temporal cell interactions that occur in vivo during spiral artery transformation, it is unclear whether trophoblast migration and endothelial tube formation as assessed in vitro validly mirror the process of UAR as it occurs in vivo. Thus, in vivo animal studies are needed to ascertain the applicability and physiological role of the candidate factors shown in vitro either operating alone or in conjunction with each other in regulating UAR.
Human clinical studies have shown that placental expression and/or maternal serum levels of many growth factors, including VEGF, IGF-I, EGF, HB-EGF, TGF, soluble endoglin, and other peptides, as well as Notch-2, endothelial colony-forming cells, tyrosine kinase-like orphan receptor, and microRNA-93, are either elevated, decreased, or unaltered in mid to late gestation in women who develop preeclampsia [23, 26, 60,61,62,63,64,65,66,67,68,69,70]. Additional studies have shown that maternal serum levels of PlGF are decreased, and the levels of the sFlt-1 soluble truncated VEGF receptor that binds to and suppresses VEGF bioavailability and endoglin were increased, preceding or coinciding with onset of the complications, e.g., maternal vascular dysfunction, of preeclampsia [71,72,73,74,75,76,77,78]. Consequently, it has been suggested that an imbalance in the levels of anti-angiogenic and angiogenic proteins and other factors may serve as biomarkers that are predictive for early detection of preeclampsia (reviewed in [19, 79, 80]).
Studies have also shown either an increase, no change, or a decrease in maternal serum estradiol levels at mid to late gestation in women exhibiting preeclampsia [81,82,83,84,85,86,87]. However, the role of estradiol in early human pregnancy with respect to UAR and onset of the pathophysiological conditions associated with preeclampsia has not been investigated.
Clinical studies have also shown that the number of immune cells, notably uNK cells, macrophages, and dendritic cells is either increased [88,89,90,91,92,93,94], decreased [95,96,97], or not altered [98,99,100] in decidua/placental bed obtained in late gestation before (e.g., biopsies) or after parturition in patients with preeclampsia. Studies also indicate that women with preeclampsia primarily express the inhibitory and not the stimulatory KIR receptors for uNK cells and that women with a KIR AA genotype, i.e., predictive of expression of the inhibitory KIRs KIR2DL-1, KIR2DL-2, KIR2DL-3, and KIR2DL-5, are at increased risk for developing preeclampsia . Macrophages are also differently activated in preeclampsia [102,103,104,105,106], which may reflect a decrease in M2 macrophages and a concomitant increase in M1 macrophages  in the placental bed of preeclamptic women. Such a change would be consistent with increased placental production of pro-inflammatory cytokines  and decreased formation of anti-inflammatory cytokines [108, 109] that occur in preeclampsia. Interestingly the levels of mRNAs for immune-associated genes, notably IL-6 and macrophages, as well as markers for expression of M2 macrophages  are greater in biopsies of decidua from women in early gestation who subsequently developed pregnancy-induced hypertension compared with those who remained normotensive.
The human studies have been important in correlating the levels of the various factors with the pathophysiological features of preeclampsia. However, it is difficult to test cause and effect and the alterations in the various factors at mid-late gestation in preeclampsia patients may result from and not underpin the pathophysiological conditions elicited by preeclampsia. Importantly, since UAR was not simultaneously examined in these clinical studies, the regulatory role of these factors on UAR has not been established in normal or adverse human pregnancy.
As presented in recent reviews [16, 19, 20, 111], early-onset preeclampsia is considered a two-stage disorder, stage 1 reflecting reduced placental perfusion and dysfunction due to impaired UAR and stage 2 the maternal syndrome induced by inadequate placental perfusion and deportation into the maternal blood of placental factors and syncytial particles produced in response to placental hypoxia and oxidative stress (Fig.1). Although the maternal disorder including organ system involvement can vary greatly in complexity and severity [16, 111], maternal systemic vascular dysfunction and hypertension are hallmark features of preeclampsia. These manifestations appear to result from vascular endothelial inflammation, oxidative stress and dysfunction, notably impaired ability to produce the vasodilators nitric oxide (NO) and prostacyclin I2, increased production of vasoconstrictors such as endothelin, and hyper-sensitivity of VSMC to vasoconstrictors within the vascular bed (Fig.2) [12, 14, 16,17,18, 20, 112,113,114,115,116,117,118,119,120,121].
The two-stage placental model of preeclampsia in which it has been proposed  that impaired remodeling of uterine spiral arteries (poor placentation) is the pathway to stage 1 preeclampsia (placental dysfunction) and the preclinical stage before development of the maternal clinical syndrome (stage 2). Reprinted from Staff 
Hypothetical scheme depicting how abnormal trophoblast invasion and spiral artery remodeling result in placental ischemia, endothelial dysfunction, and hypertension in preeclampsia. Reprinted from Palei et al. 
It is well established that VEGF plays a pivotal role in promoting vascular endothelial cell integrity, stability, and function, including NO production [122,123,124]. Therefore, it has been hypothesized [19, 75, 117] that in preeclampsia the placental ischemia induced by defective UAR causes an increase in placental expression and maternal serum levels of sFlt-1, which decreases VEGF bioavailability and elicits maternal vascular dysfunction (Fig. 2). Accordingly, animal models have been developed to examine this hypothesis and ascertain the possibility of achieving a therapeutic approach to overcome or prevent the vascular dysfunction elicited by decreased bioavailability of VEGF. Thus, key manifestations of preeclampsia, i.e., maternal hypertension, fetal growth restriction, and/or maternal vascular endothelial dysfunction, were induced in mice or rats in which levels of sFlt1 and/or endoglin were elevated by systemic adenoviral delivery of these proteins [74, 125,126,127,128,129,130,131]. Systemic administration to mice of an antibody which neutralized both Flt-1 and sFlt-1 decreased uterine artery length as an index of arterial transformation in this species . The clinical manifestations of preeclampsia elicited in several of these animal models were prevented by adenoviral delivery of VEGF121 [133,134,135,136,137,138]. Symptoms of preeclampsia were also overcome in lentiviral sFlt1-treated mice by concomitant administration of the drug pravastatin  and in BPH/5 mice by injection of the drug celecoxib at the time of embryo implantation, which apparently acted by restoring the levels of Cox 2, VEGF, and related angiogenic factors . Mice defective for PlGF, a member of the VEGF family, also exhibit preeclampsia-like symptoms, notably maternal endothelial dysfunction, as well as cognitive function of the offspring [141, 142]. Moreover, an experimental increase in sFlt-1 levels or decrease in VEGF and PlGF levels induced in rats and sheep by aortic or uterine artery ligation to elicit placental ischemia caused maternal hypertension, proteinurea, and vascular dysfunction, effects reversed by VEGF or PlGF administration [135, 138, 143,144,145]. Uterine spiral arteriole remodeling and MMP-2 and MMP-9 were decreased in the rat reduced uterine perfusion pressure model .
Inhibition of NO synthase , as well as administration of tumor necrosis factor-  or interleukin , also induced preeclampsia-like symptoms in mice and rats. Interestingly, uterine artery diameter and length were reduced in endothelial NO synthase-null mice , whereas nanoparticle-mediated delivery of the NO donor SE175 to mice at mid-late gestation increased spiral artery diameter . Roles for the Notch signaling pathway and the storkhead box 1 (STOX 1) transcription factor have also been suggested since Notch 2-null mice exhibited a decrease in spiral artery diameter and placental perfusion , while overexpression of STOX 1 in mice led to a preeclampsia phenotype of hypertension .
The role of immune cells in the process of vessel transformation has been proposed. Thus, studies in uNK cell-immunodeficient mice indicate that uNK cells, via the formation of interferon gamma, promote modification (i.e., luminal area) of spiral arteries [153,154,155,156]. Moreover, T lymphocyte regulatory cell (Tregs)-deficient mice show impaired uterine artery remodeling and flow [157, 158], suggesting that Tregs impair inflammatory responses that cause a defect in uterine vessel transformation .
Evidence for involvement of the renin-angiotensin (AT)-aldosterone system in preeclampsia has also emerged from rodent models. Thus, administration of antibodies to AT1 beginning at midgestation to mice or rats elicited hypertension, proteinuria, glomerular endotheliosis, and placental abnormalities [160, 161]. Moreover, AT1-deficient mice exhibited impaired placentation , and angiotensinogen transgenic mice exhibited deeper endovascular trophoblast invasion and spiral artery remodeling . Upregulation of VSMC AT1 expression elicited hypertension, proteinuria, increased sFlt-1 expression, and decreased placental labyrinth growth in mice, effects prevented by administration of -arrestin, a G protein that causes AT1 receptor desensitization .
The rodent models have been valuable in recapitulating the clinical symptoms of pregnancy disorders such as preeclampsia. However, in most instances, UAR was not examined, experimental interventions used to induce preeclampsia-like symptoms were often applied after the time of placentation, and many of the clinical features of preeclampsia were also induced in nonpregnant rodents, and thus, these models were not specific for pregnancy. Moreover, there are significant differences in placental morphology and development, the process and impact of spiral artery remodeling, uterine and placental vascular anatomy, and the maternal-placenta-fetal endocrine inter-relationships between rodents and humans [58, 59, 165,166,167,168,169,170,171]. For example, in the mouse and rat, trophoblast invasion is temporally restricted to late gestation [58, 172] and the role of UAR on maternal vascular function may be equivocal. Thus, although NK-defective mice exhibit impaired UAR, maternal resting blood pressure remains normal throughout gestation and maternal proteinuria does not develop , while trophoblast arterial invasion is more extensive and uterine artery resistance lower in the rat BHP/5 model of preeclampsia . Collectively, these differences between rodents and humans make translation of findings on UAR in the rodent to the human uncertain.
Although rodents have been extensively used to recapitulate the pathophysiological features of preeclampsia, relatively few studies have employed nonhuman primates in this area of perinatal biology. Placental morphology, the process of uterine spiral artery transformation, uterine and placental vascular anatomy, and maternal-placental-fetal endocrine inter-relationships are similar in human and baboon pregnancy [58, 165, 173]. Although remodeling of the spiral arteries in the baboon does not extend into the inner myometrium, as in human pregnancy, the qualitative nature of placentation and UAR are alike [58, 174]. In addition to these important considerations, humans and baboons exhibit similar anatomy, physiology, and ontogeny of the fetal-placental unit  and share >96% DNA/genetic homology [175, 176], and thus, the baboon provides an excellent nonhuman primate model for the study of human placental and fetal development.
As in the rodent studies, uterine artery ligation has been employed as an experimental paradigm in pregnant baboons. Uteroplacental ischemia elicited by uterine artery ligation in the second half of baboon pregnancy resulted in hypertension, proteinuria, and renal endotheliosis, effects reversed by administration of sFlt-1 siRNA or PlGF [177,178,179,180]. Thus, the latter primate studies focused on recapitulating the symptoms of adverse human pregnancy but not on UAR.
In contrast to the latter approach, the authors have published a series of studies in which they have established an experimental paradigm of prematurely elevating estradiol levels in the first trimester of baboon pregnancy to study the regulation of UAR [181,182,183]. Slightly elevating maternal estradiol levels resulted in a 3-fold increase in placental expression and maternal serum levels of sFlt-1 and decrease in extravillous trophoblast expression of VEGF in early pregnancy (Fig.3). The increase in sFlt-1/decrease in VEGF was associated with a 75% reduction in the level of UAR, quantified as the percent of uterine spiral arteries invaded and remodeled by extravillous trophoblasts, at the end of the first trimester (Fig.4). Concomitant administration of estradiol and delivery of the VEGF gene selectively to the maternal aspect of the placenta, but not the fetus, by contrast-enhanced ultrasonography/microbubble technology restored VEGF protein levels and prevented the decrease in UAR (Fig. 4, ).
(a) sFlt-1 levels in uterine vein and (b) VEGF protein quantified by proximity ligation assay (signals/nuclear area 104) in the anchoring villi on day 60 in untreated and estradiol (E2)-treated baboons. *P<0.05
Percent remodeling of uterine spiral arteries (i.e., number of vessels exhibiting trophoblast invasion divided by total number of vessels counted) on day 60 of gestation in baboons untreated, treated with estradiol (E2), or treated with E2 plus VEGF DNA. *Different (P<0.01) from values in other two groups
The decline in extravillous trophoblast VEGF expression in estradiol-treated baboons was associated with a decrease in expression of the 11 and 51 integrins  that promote trophoblast migration and remodeling and are increased by VEGF in vitro [185,186,187]. This suggests that these integrins may mediate the stimulatory effect of VEGF on UAR during early baboon pregnancy. Coinciding with the decrease in UAR, uterine artery blood flow was reduced by 30% and maternal blood pressure increased by 25% near term, suggesting an impairment of maternal systemic vascular function . Although it has been suggested that the alteration in expression of pro- and anti-angiogenic growth factors is simply the result and not the cause of placental dysfunction in preeclampsia , the prevention of the decrease in UAR by VEGF delivery in early baboon pregnancy is consistent with VEGF having a pivotal role in promoting UAR.
UAR is vital to successful pregnancy; however, the regulation of this fundamentally important process has not been established. The in vitro studies are important in having identified a multitude of factors that have the ability to alter migratory and invasive capacity of trophoblast cells. However, it is unclear whether trophoblast migration and endothelial tube formation as assessed in vitro validly mirror the process of UAR as it occurs in vivo. The clinical studies have been significant in showing that maternal serum levels of certain factors are altered, particularly sFlt-1 which is increased and PlGF which is decreased, preceding or coinciding with onset of the complications, e.g., maternal vascular dysfunction, emanating from preeclampsia. However, it is difficult to test cause and effect in human pregnancy studies, and thus, the alteration in circulating levels of the various factors may be a consequence of and not underpin the pathophysiological conditions elicited by adverse pregnancy. The rodent and a few primate studies have been valuable in recapitulating, and showing the ability of certain growth factors to mitigate, the clinical manifestations of pregnancy disorders such as preeclampsia, but have not focused on UAR. This review has described the results of in vitro, clinical, and rodent studies and also a novel experimental model of defective UAR in a nonhuman primate that allows study of the regulation of spiral artery transformation and the potential to develop therapeutic modalities to manage or prevent the maternal pathophysiological consequences of adverse pregnancy arising from defective UAR.
Khong TY, De Wolf F, Robertson WB, Brosens I. Inadequate maternal vascular response to placentation in pregnancies complicated by pre-eclampsia and by small-for-gestational age infants. Br J Obstet Gynaecol. 1986;93:104959.
Gilbert JS, Ryan MJ, LaMarca BB, Sedeek M, Murphy SR, Granger JP. Pathophysiology of hypertension during preeclampsia: linking placental ischemia with endothelial dysfunction. Am J Physiol Heart Circ Physiol. 2008;294:H54150.
Hannon T, Innes BA, Lash GE, Bulmer JN, Robson SC. Effects of local decidua on trophoblast invasion and spiral artery remodeling in focal placenta creta an immunohistochemical study. Placenta. 2012;33:9981004.
Zhou Y, McMaster M, Woo K, Janatpour M, Perry J, Karpanen T, et al. Vascular endothelial growth factor ligands and receptors that regulate human cytotrophoblast survival are dysregulated in severe preeclampsia and hemolysis, elevated liver enzymes, and low platelets syndrome. Am J Pathol. 2002;160:140523.
Anteby EY, Greenfield C, Natanson-Yaron S, Goldman-Wohl D, Hamani Y, Khudyak V, et al. Vascular endothelial growth factor, epidermal growth factor and fibroblast growth factor-4 and -10 stimulate trophoblast plasminogen activator system and metalloproteinase-9. Mol Hum Reprod. 2004;10:22935.
Leach RE, Kilburn B, Wang J, Liu Z, Romero R, Armant DR. Heparin-binding EGF-like growth factor regulates human extravillous cytotrophoblast development during conversion to the invasive phenotype. Dev Biol. 2004;266:22337.
Lala N, Girish GV, Cloutier-Bosworth A, Lala PK. Mechanisms in decorin regulation of vascular endothelial growth factor-induced human trophoblast migration and acquisition of endothelial phenotype. Biol Reprod. 2012;87(59):114.
Evans J, Salamonsen LA, Menkhorst E, Dimitradis E. Dynamic changes in hyperglycosylated human chorionic gonadotrophin throughout the first trimester of pregnancy and its role in early placentation. Hum Reprod. 2015;30:102938.
Fitzpatrick TE, Lash GE, Yanaihara A, Charnock-Jones DS, Macdonald-Goodfellow SK, Graham CH. Inhibition of breast carcinoma and trophoblast cell invasiveness by vascular endothelial growth factor. Exp Cell Res. 2003;283:24755.
Jones RL, Findlay JK, Farnworth PG, Robertson DM, Wallace E, Salamonsen LA. Activin A and inhibin A differentially regulate human uterine matrix metalloptoteinases: potential interactions during decidualization and trophoblast invasion. Endocrinology. 2006;147:72432.
Tamaru S, Mizuno Y, Tochigi H, Kajihara T, Okazaki Y, Okagaki R, et al. MicroRNA-135b suppresses extravillous trophoblast-derived HTR-7/SVneo cell invasion by directly down regulating CXCL12 under low oxygen conditions. Biochem Biophys Res Commun. 2015;461:4216.
Caniggia I, Grisaru-Gravnosky S, Kuliszewsky M, Post M, Lye SJ. Inhibition of TFG-beta 3 restores the invasive capability of extravillous trophoblasts in preeclamptic pregnancies. J Clin Invest. 1999;103:164150.
Majali-Martinez A, Velicky P, Pollheimer J, Knfler M, Yung HW, Burton GJ, et al. Endothelin-1 down-regulates matrix metalloproteinase 14 and 15 expression in human first trimester trophoblasts via endothelin receptor type B. Hum Reprod. 2017;32:4654.
Choudhury RH, Dunk CE, Lye SJ, Aplin JD, Harris LK, Jones RL. Extravillous trophoblast and endothelial cell crosstalk mediates leukocyte infiltration to the early remodeling decidual spiral arteriole wall. J Immunol. 2017;198(10):411528.
Sharkey AM, Cooper JC, Balmforth JR, McLaren J, Clark DE, Charnock-Jones DS, et al. Maternal plasma levels of vascular endothelial growth factor in normotensive pregnancies and in pregnancies complicated by pre-eclampsia. Eur J Clin Investig. 1996;26:11825.
Chung JY, Song Y, Wang Y, Magness RR, Zheng J. Differential expression of vascular endothelial growth factor (VEGF), endocrine gland derived-VEGF, and VEGF receptors in human placentas from normal and preeclamptic pregnancies. J Clin Endocrinol Metab. 2004;89:248490.
Hamilton GS, Lysiak JJ, Han VK, Lala PK. Autocrine-paracrine regulation of human trophoblast invasiveness by insulin-like growth factor (IGF)-II and IGF-binding protein (IGFBP)-1. Exp Cell Res. 1998;244:14756.
Fragkiadaki P, Soulitzis N, Sifakis S, Koutroulakis D, Gourvas V, Vrachnis N, et al. Downregulation of notch signaling pathway in late preterm and term placentas from pregnancies complicated by preeclampsia. PLoS One. 2015;10:e0126163.
Muoz-Hernandez R, Miranda ML, Stiefel P, Lin RZ, Praena-Fernndez JM, Dominguez-Simeon MJ, et al. Decreased level of cord blood circulating endothelial colony-forming cells in preeclampsia. Hypertension. 2014;64:16571.
Travaglino A, Raffone A, Saccone G, Migliorini S, Maruotti GM, Esposito G, et al. Placental morphology, apoptosis, angiogenesis and epithelial mechanisms in early-onset preeclampsia. Eur J Obstet Gynecol Reprod Biol. 2019;234:2006.
Chen J, Yue C, Xu J, Zhan Y, Zhao H, Li Y, et al. Downregulation of receptor tyrosine kinase-like orphan receptor 1 in preeclampsia placenta inhibits human trophoblast cell proliferation, migration, and invasion by PI3K/AKT/mTOR pathway accommodation. Placenta. 2019;82:1724.
Taylor RN, Grimwood J, Taylor RS, McMaster MT, Fisher SJ, North RA. Longitudinal serum concentrations of placental growth factor: evidence for abnormal placental angiogenesis in pathologic pregnancies. Am J Obstet Gynecol. 2003;188:17782.
Tsatsaris V, Goffin F, Munaut C, Brichant JF, Pignon MR, Noel A, et al. Overexpression of the soluble vascular endothelial growth factor receptor in preeclamptic patients: pathophysiological consequences. J Clin Endocrinol Metab. 2003;88:555563.
Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S, et al. Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest. 2003;111:64958.
Chaiworapongsa T, Romero R, Kim YM, Kim GJ, Kim MR, Espinoza J, et al. Plasma soluble vascular endothelial growth factor receptor-1 concentration is elevated prior to the clinical diagnosis of pre-eclampsia. J Matern Fetal Neonatal Med. 2005;17:318.
Wathn KA, Tuutti E, Stenman UH, Alfthan H, Halmesmki E, Finne P, et al. Maternal serum-soluble vascular endothelial growth factor receptor-1 in early pregnancy ending in preeclampsia or intrauterine growth retardation. J Clin Endocrinol Metab. 2006;91:1804.
Kenny LC, Black MA, Poston L, Taylor R, Myers JE, Baker PN, et al. Early pregnancy prediction of preeclampsia in nulliparous women, combining clinical risk and biomarkers: the screening for pregnancy endpoints (SCOPE) international cohort study. Hypertension. 2014;64:64452.
Serin IS, Kula M, Babu M, Unlhizarci K, Ger S, Tayyar M. Androgen levels of preeclamptic patients in the third trimester of pregnancy and six weeks after delivery. Acta Obstet Gynecol Scand. 2001;80:100913.
Jobe SO, Tyler CT, Magness RR. Aberrant synthesis, metabolism, and plasma accumulation of circulating estrogens and estrogen metabolites in preeclampsia implications for vascular dysfunction. Hypertension. 2013;61:4807.
Bachmayer N, Rafik Hamad R, Liszka L, Bremme K, Sverremark-Ekstrom E. Aberrant uterine natural killer (NK)-cell expression and altered placental and serum levels of the NK-cell promoting cytokine interleukin-12 in pre-eclampsia. Am J Reprod Immunol. 2006;56(56):292301.
Wilczyski JR, Tchrzewski H, Banasik M, Gowacka E, Wieczorek A, Lewkowicz P, et al. Lymphocyte subset distribution and cytokine secretion in third trimester decidua in normal pregnancy and preeclampsia. Eur J Obstet Gynecol Reprod Biol. 2003;109(1):815.
Schonkeren D, van der Hoorn ML, Khedoe P, Swings G, van Beelen E, Claas F, et al. Differential distribution and phenotype of decidual macrophages in preeclamptic versus control pregnancies. Am J Pathol. 2011;178(2):70917.
Reister F, Frank HG, Kingdom JC, Heyl W, Kaufmann P, Rath W, et al. Macrophage-induced apoptosis limits endovascular trophoblast invasion in the uterine wall of preeclamptic women. Lab Investig. 2001;81(8):114352.
Kim JS, Romero R, Cushenberry E, Kim YM, Erez O, Nien JK, et al. Distribution of CD14+ and CD68+ macrophages in the placental bed and basal plate of women with preeclampsia and preterm labor. Placenta. 2007;28(56):5716.
Williams PJ, Bulmer JN, Searle RF, Innes BA, Robson SC. Altered decidual leucocyte populations in the placental bed in pre-eclampsia and foetal growth restriction: a comparison with late normal pregnancy. Reproduction. 2009;138(1):17784.
Eide IP, Rolfseng T, Isaksen CV, Mecsei R, Roald B, Lydersen S, et al. Serious foetal growth restriction is associated with reduced proportions of natural killer cells in decidua basalis. Virchows Arch. 2006;448(3):26976.
Rieger L, Segerer S, Bernar T, Kapp M, Majic M, Morr AK, et al. Specific subsets of immune cells in human decidua differ between normal pregnancy and preeclampsiaa prospective observational study. Reprod Bio Endocrino: RBE (Rev Bras Entomol). 2009;7:132.
Hsu P, Santner-Nanan B, Dahlstrom JE, Fadia M, Chandra A, Peek M, et al. Altered decidual DC-SIGN+ antigen-presenting cells and impaired regulatory T-cell induction in preeclampsia. Am J Pathol. 2012;181(6):214960.
Hiby SE, Walker JJ, OShaughnessy KM, Redman CW, Carrington M, Trowsdale J, et al. Combinations of maternal KIR and fetal HLA-C genes influence the risk of preeclampsia and reproductive success. J Exp Med. 2004;200(8):95765.
Reister F, Frank HG, Heyl W, Kosanke G, Huppertz B, Schrder W, et al. The distribution of macrophages in spiral arteries of the placental bed in pre-clampsia differs from that in healthy patients. Placenta. 1999;20((23)):22933.
Katabuchi H, Yih S, Ohba T, Matsui K, Takahashi K, Takeya M, et al. Characterization of macrophages in the decidual atherotic spiral artery with special reference to the cytology of foam cells. Med Electron Microsc. 2003;36(4):25362.
Haeger M, Unander M, Norder-Hansson B, Tylman M, Bengtsson A. Complement, neutrophil, and macrophage activation in women with severe preeclampsia and the syndrome of hemolysis, elevated liver enzymes, and low platelet count. Obstet Gynecol. 1992;79(1):1926.
Rein DT, Breidenbach M, Hnscheid B, Friebe-Hoffmann U, Engel H, Ghring UJ, et al. Preeclamptic women are deficient of interleukin-10 as assessed by cytokine release of trophoblast cells in vitro. Cytokine. 2003;23(45):11925.
Prins JR, Faas MM, Melgert BN, Huitema S, Timmer A, Hylkema MN, et al. Altered expression of immune-associated genes in first-trimester human decidua of pregnancies later complicated with hypertension or foetal growth restriction. Placenta. 2012;33(5):4535.
Barden A, Beilin LJ, Ritchie J, Walters BN, Michael CA. Plasma and urinary endothelin 1, prostacyclin metabolites and platelet consumption in pre-eclampsia and essential hypertensive pregnancy. Blood Press. 1994;3:3846.
Cunningham FG, Roberts JM, Lindheimer MD. The clinical spectrum of preeclampsia. In: Lindheimer MD, Roberts JM, Cunningham FG, editors. Chesleys hypertensive disorders in pregnancy. San Diego: Academic; 2009. p. 2536.
Ilekis JV, Tsilou E, Fisher S, Abrahams VM, Soares MJ, Cross JC, et al. Placental origins of adverse pregnancy outcomes: potential molecular targets: an Executive Workshop Summary of the Eunice Kennedy Shriver National Institute of Child Health and Human Development. Am J Obstet Gynecol. 2016;215:S1S46.
Lu F, Longo M, Tamayo E, Maner W, Al-Hendy A, Anderson GD, et al. The effect of over-expression of sFlt-1 on blood pressure and the occurrence of other manifestations of preeclampsia in unrestrained conscious pregnant mice. Am J Obstet Gynecol. 2007;196:396.e17.
Tang JR, Karumanchi SA, Seedorf G, Markham N, Abman SH. Excess soluble vascular endothelial growth factor receptor-1 in amniotic fluid impairs lung growth in rats: linking preeclampsia with bronchopulmonary dysplasia. Am J Physiol Lung Cell Mol Physiol. 2012;302:L3646.
Suzuki H, Ohkuchi A, Matsubara S, Takei Y, Murakami M, Shibuyua M, et al. Effect of recombinant placental growth factor 2 on hypertension induced by full-length mouse soluble fms-like tyrosine kinase 1 adenoviral vector in pregnant mice. Hypertension. 2009;54:11291135.
Bridges JP, Gilbert JS, Colson D, Gilbert SA, Dukes MP, Ryan MJ, et al. Oxidative stress contributes to soluble fms-like tyrosine kinase-1 induced vascular dysfunction in pregnant rats. Am J Hypertens. 2009;22:5648 Epub 2009 Mar 5.
Szalai G, Romero R, Chaiworapongsa T, Xu Y, Wang B, Ahn H, et al. Full-length human placental sFlt-1-e15a isoform induces distinct maternal phenotypes of preeclampsia in mice. PLoS One. 2015;10(4):e0119547.
Khankin EV, Mandala M, Colton I, Karumanchi SA, Osol G. Hemodynamic, vascular, and reproductive impact of FMS-like tyrosine kinase 1 (FLT1) blockade on the uteroplacental circulation during normal mouse pregnancy. Biol Reprod. 2012;86:57,18.
Li Z, Zhang Y, Ying Ma J, Kapoun AM, Shao Q, Kerr I, et al. Recombinant vascular endothelial growth factor 121 attenuates hypertension and improves kidney damage in a rat model of preeclampsia. Hypertension. 2007;50:68692.
Bergmann A, Ahmad S, Cudmore M, Gruber AD, Wittschen P, Lindenmaier W, et al. Reduction of circulating soluble Flt-1 alleviates preeclampsia-like symptoms in a mouse model. J Cell Mol Med. 2010;14:185767.
Gilbert JS, Verzwyvelt J, Colson D, Arany M, Karumanchi SA, Granger JP. Recombinant vascular endothelial growth factor 121 infusion lowers blood pressure and improves renal function in rats with placental ischemia-induced hypertension. Hypertension. 2010;55:380385.
Woods AK, Hoffmann DS, Weydert CJ, Butler SD, Zhou Y, Sharma RV, et al. Adenoviral delivery of VEGF121 early in pregnancy prevents spontaneous development of preeclampsia in BPH/5 mice. Hypertension. 2011;57:94102.
Mateus J, Bytautiene E, Lu F, Tamayo EH, Betancourt A, Hankins GD, et al. Endothelial growth factor therapy improves preeclampsia-like manifestations in a murine model induced by overexpression of sVEGFR-1. Am J Physiol Heart Circ Physiol. 2011;301:H17817.
Kumasawa K, Ikawa M, Kidoya H, Hasuwa H, Saito-Fujita T, Morioka Y, et al. Pravastatin induces placental growth factor (PGF) and ameliorates preeclampsia in a mouse model. Proc Natl Acad Sci U S A. 2011;108:14515.
Rejinders D, Liu CC, Xu X, Zhao AM, Olson KN, Butler SD, et al. Celecoxib restores angioogenic factor expression at the maternal-fetal interface in the BPH/5 mouse model of preeclampsia. Physiol Genomics. 2018;50(5):38592.
LaMarca BB, Bennett WA, Alexander BT, Cockrell K, Granger JP. Hypertension produced by reductions in uterine perfusion in the pregnant rat: role of tumor necrosis factor. Hypertension. 2005;46:10225.
Gilbert JS, Babcock SA, Granger JP. Hypertension produced by reduced uterine perfusion in pregnant rats is associated with increased soluble fms-like tyrosine kinase-1 expression. Hypertension. 2007;50:11427 Epub 2007 Oct 8.
Walsh SK, English FA, Johns EJ, Kenny LC. Plasma-mediated vascular dysfunction in the reduced uterine perfusion pressure model of preeclampsia: a microvascular characterization. Hypertension. 2008;54:34551.
Lin, C, He H, Cui N, Ren Z, Zhu M, Khalil RA. Decreased uterine vascularization and uterine arterial expansive remodeling with reduced matrix metalloproteinase-2 and -9 in hypertensive pregnancy. Am J Physiol Heart Circ Physiol. 2019;318(1):H165H180.
Molnr M, St T, Tth T, Hertelendy F. Prolonged blockade of nitric oxide synthesis in gravid rats produces sustained hypertension, proteinuria, thrombocytopenia, and intrauterine growth retardation. Am J Obstet Gynecol. 1994;170:145866.
Kulandavelu S, Whiteley KJ, Qu D, Mu J, Bainbridge SA, Adamson SL. Endothelial nitric oxide synthase deficiency reduces uterine blood flow, spiral artery elongation, and placental oxygenation in pregnant mice. Hypertension. 2012;60:2318.
Cureton N, Kortokova I, Baker B, Greenwood S, Wareing M, Kotamraju VR, et al. Selective targeting of a novel vasodilator to the uterine vasculature to treat impaired uteroplacental perfusion in pregnancy. Theronastics. 2017;7(15):371531 eCollection 2017.
Hunkapillar NM, Gasperowicz M, Kapidzic M, Plaks V, Maltepe E, Kitajewski J, et al. A role for Notch signaling in trophoblast endovascular invasion and in the pathogenesis of pre-eclampsia. Development. 2011;138:298798.
Doridot L, Passet B, Mhats C, Rigourd V, Barbaux S, Ducat A, et al. Preeclampsia-like symptoms induced in mice by fetoplacental expression of STOX1 are reversed by aspirin treatment. Hypertension. 2013;61:6628.
Ashkar AA, Di Santo JP, Croy BA. Interferon gamma contributes to initiation of uterine vascular modification, decidual integrity, and uterine natural killer cell maturation during normal murine pregnancy. J Exp Med. 2000;192:25970.
Nadkarni S, Smith J, Sferruzzi-Perri AN, Ledwozyw A, Kishore M, Haas R, et al. Neutrophils induce proangiogenic T cells with a regulatory phenotype in pregnancy. Proc Natl Acad Sci U S A. 2016;113:E841524.
Parrish MR, Murphy SP, Rutland S, Wallace K, Wenzel K, Wallukat G, et al. The effect of immune factors, tumor necrosis factor-alpha, and agonistic autoantibodies to the angiotensin II type I receptor on soluble fms-like tyrosine-1 and soluble endoglin production in response to hypertension during pregnancy. Am J Hypertens. 2010;23:9116 Epub 2010 Apr 29.
Geusens N, Hering L, Verlohren S, Luyten C, Drijkoningen K, Taube M, et al. Changes in endovascular trophoblast invasion and spiral artery remodelling at term in a transgenic preeclamptic rat model. Placenta. 2010;31:3206.
Adamson SL, Lu Y, Whiteley KJ, Homyard D, Hemberger M, Pfarrer C, et al. Interactions between trophoblast cells and the maternal and fetal circulation in the mouse placenta. Dev Biol. 2002;250(2):35873.
Schlabritz-Loutsevitch NE, Hubbard GB, Jenkins SL, Martin HC, Snider CS, Frost PA, et al. Ontogeny of hematological cell and biochemical profiles in maternal and fetal baboons (Papio species). J Med Primatol. 2005;34:193200.
Makris A, Yeung KR, Lim SM, Sunderland N, Heffernan S, Thompson JF, et al. Placental growth factor reduces blood pressure in a uteroplacental ischemia model of preeclampsia in nonhuman primates. Hypertension. 2016;67:126372.
Bonagura TW, Pepe GJ, Enders AC, Albrecht ED. Suppression of extravillous trophoblast vascular endothelial growth factor expression and uterine spiral artery invasion by estrogen during early baboon pregnancy. Endocrinology. 2008;149:507887.
Bonagura TW, Babischkin JS, Aberdeen GW, Pepe GJ, Albrecht ED. Prematurely elevating estradiol in early baboon pregnancy suppresses uterine artery remodeling and expression of extravillous placental vascular endothelial growth factor and 11 and 51 integrins. Endocrinology. 2012;153:2897906.
Aberdeen GW, Bonagura TW, Harman CR, Pepe GJ, Albrecht ED. Suppression of trophoblast uterine spiral artery remodeling by estrogen during baboon pregnancy: impact on uterine and fetal blood flow dynamics. Am J Physiol Heart Circ Physiol. 2012;302:H193644.
Babischkin JS, Aberdeen GW, Lindner JR, Bonagura TW, Pepe GJ, Albrecht ED. Vascular endothelial growth factor delivery to placental basal plate promotes uterine artery remodeling in the primate. Endocrinology. 2019;160:1492505.
Aplin JD, Haigh T, Jones CJ, Church HJ, Viovac L. Development of cytrophoblast columns from explanted first trimester human placental villi: role of fibronectin and integrin 51. Biol Reprod. 1999;60:828838.
Fukushima K, Miyamoto S, Tsukimori K, Kobayashi H, Seki H, Takeda S, et al. Tumor necrosis factor and vascular endothelial growth factor induce endothelial integrin repertories, regulating endovascular differentiation and apoptosis in a human extravillous trophoblast cell line. Biol Reprod. 2005;73:1729.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Ischemic placental disease is a concept that links intrauterine growth retardation (IUGR) and preeclampsia (PE) back to insufficient remodeling of uterine spiral arteries. The rheological consequences of insufficient remodeling of uterine spiral arteries were hypothesized to mediate the considerably later manifestation of obstetric disease. However, the micro-rheology in the intervillous space (IVS) cannot be examined clinically and rheological animal models of the human IVS do not exist. Thus, an in silico approach was implemented to provide in vivo inaccessible data. The morphology of a spiral artery and the inflow region of the IVS were three-dimensionally reconstructed to provide a morphological stage for the simulations. Advanced high-end supercomputing resources were used to provide blood flow simulations at high spatial resolution. Our simulations revealed turbulent blood flow (high-velocity jets and vortices) combined with elevated blood pressure in the IVS and increased wall shear stress at the villous surface in conjunction with insufficient spiral artery remodeling only. Post-hoc histological analysis of uterine veins showed evidence of increased trophoblast shedding in an IUGR placenta. Our data support that rheological alteration in the IVS is a relevant mechanism linking ischemic placental disease to altered structural integrity and function of the placenta.
Ischemic placental disease is considered to be a unifying pathogenetic concept that connects various obstetric syndromes, particularly intrauterine growth retardation (IUGR) and preeclampsia (PE)1, which are the major causes of iatrogenic preterm birth2,3,4,5. The concept of ischemic placental disease traces the roots of IUGR and PE back to an early common pathogenetic origin, i.e., insufficient dilation of uterine spiral arteries6,7,8. Dilation of uterine spiral arteries is normally accomplished through remodeling of the arterial media and replacement of endothelium by invading fetal trophoblast cells9,10 during the first trimester of pregnancy, a phase of predominantly histiotrophic nutrition7,8. Clinically symptomatic manifestation of the syndromes IUGR and PE occurs beyond the 20th week of gestation4,7,11, after the switch to hematotrophic nutrition.
Those elements of the pathogenetic chain postulated by the concept of ischemic placental disease that range from the early cause to the late onset of symptoms are not fully understood, although the actually proposed mechanisms are rheological by nature12. Investigating the structure of the uterine spiral arteries at the microscopic scale and their function in vivo is not possible. Furthermore, animal models do not exist in this regard due to substantial species differences in implantation, degree and depth of trophoblast invasion, arterial remodeling and barrier morphology13. Blood flow simulations of the uteroplacental circulation are thus a promising way to approach the mechanisms of development of symptomatic diseases beyond the 20th week of gestation. Unfortunately, the resolution limits of ultrasound imaging are also limiting the spatial resolution of such computation models in which both morphological data and flow data were derived from ultrasound examinations14,15.
A recent review of this problem included static calculations of blood flow through straight tubes with or without conical tube openings (as models of remodeled or unremodeled spiral arteries, respectively), which were based on the Hagen-Poiseuille equation12. The calculations performed in the latter study used static flow and pressure conditions and included literature-deduced assumptions of, e.g., uterine artery total flow, numbers of spiral arteries and diameters of arteries with normally dilated and undilated openings. Although this study12 indicated the promise of blood flow simulations to explain otherwise inaccessible in vivo conditions in human pregnancy, the implemented model still possessed numerous inherent limitations that compromised its proximity to real blood flow conditions in utero and thus also its relevance. To achieve the best possible proximity to the in vivo situation and to increase relevance, a novel model will have to implement several thus far missing important aspects: (i) rather than straight tubes, spiralized arterial segments similar to those in vivo should be included in the model environment; (ii) the intervillous space (IVS) proximal to the arterial opening and the villous tree in the area proximal to the arterial opening have to be included in the calculations; (iii) rheological models will have to exceed the limitations of the Hagen-Poiseuille equation (primarily the restricted validity of the Hagen-Poiseuille equation for laminar flow in tubes only); (iv) modeling algorithms should be able to deliver data on flow velocities, pressure and wall shear stress at vessel walls and at the surface of the villous tree with sufficient spatial resolution in all parts of the model environment, as well as in regions that cannot be considered as tubes (e.g., the IVS or the surface of the villous tree); and (v) the flow model should include flow and pressure fluctuations in the entire model environment during at least one complete maternal heart cycle in the clinically normal, IUGR and IUGR combined with PE (IUGR/PE) conditions.
The model of the present study was designed to meet the aforementioned requirements. It was constructed on a morphological stage with a size of 300025002000m covering the inflow region of a spiral artery, including a substantial spiralized segment of this artery and the first 2000m of IVS and villous tree proximal to the opening of the spiral artery (proximal IVS, see Figure S2 for topology definition). The region was three-dimensionally reconstructed from serial histological sections. The blood flow simulations used an innovative approach with a full description of blood flow properties (velocity profile, pressure profile and wall shear stress profile) at high resolution (550443120 voxels) in the entire morphological stage of the model. Exemplary Doppler flow data of uterine arteries were obtained from a clinically normal patient and from patients with IUGR and IUGR/PE to feed the simulations throughout three full maternal heart cycles.
The present study demonstrates velocity jets projecting into the proximal IVS which correspond to elevated wall shear stress at the villous surfaces in the proximal IVS. Vortices and elevated blood pressure were predicted in the proximal IVS. Circular vortices close to the unremodeled arterial openings are evidence of dead volumes of circling blood, which could compromise the efficiency of nutrient exchange. Furthermore, using an innovative, non-routine histological approach, this study also provides evidence that increased wall shear stress can indeed correspond to increased trophoblast shedding from the stressed villous surface.
The three simulations showed similar flow velocities at the inlets of the arteries, corresponding to the Doppler flow data. In the clinically normal situation, the velocity decreased at the arterial opening and the blood flow remained slow throughout the proximal IVS (Fig. 1A). The pathological cases (Fig. 1B,C), however, showed high-velocity jets in the entry region into the proximal IVS and a snap-through into the more distant regions. Jet streams started after 0.55sec of the pulse cycle at a velocity of 70cm/s and showed peak velocities that were a factor of 5 higher in IUGR and a factor of 4 higher in IUGR/PE compared to the clinically normal situation.
(A and D) show the model of the clinically normal situation with a dilated arterial opening, (B and E) the model of the situation in IUGR, and (C and F) the model of the situation in IUGR/PE. (AF) show simulated blood flow at the point of maximum systolic flow of a maternal heart cycle. (AC) show blood velocity vectors, which are color-coded from low (blue) to high (red) velocity; the velocity color scale is shown in C (right to the proximal IVS). The proximal IVS cuboids of (AC) contain blue planes that mark the distance from the arterial opening (scale shown in (A) right of the proximal IVS). In (A) the velocity decreases at the arterial opening, with dominating blue colors at the arterial opening. In (B) and less pronounced in (C) the velocity accelerates at the arterial opening, forming a velocity jet that projects deeply into the proximal IVS. (DF) show wall shear stress; the color scale for (DF) (in Pa) is shown in (D) (left of the proximal IVS). In the clinically normal situation (D), there are moderate to low values of wall shear stress at the villous surface in the proximal IVS. In (E) and less pronounced in (F) the wall shear stress is elevated in central parts of the proximal IVS, as indicated by more red colors appearing at the villous surface.
The computed wall shear stresses at the surfaces of the villous trees in the proximal IVS were low in the model of clinically normal pregnancy (Fig. 1D). In contrast, both pathological conditions showed elevated levels of wall shear stress at the surfaces of the villous trees in the proximal IVS, which were slightly more pronounced in IUGR than in IUGR/PE at an intermediate distance from the arterial opening (Fig. 1E,F). The maximum values for computed wall shear stresses are listed in detail in Table S2, and the rise of wall shear stress during the maternal heart cycle is shown in the movies (Movie S1a for clinically normal and Movie S1b for IUGR) in Supplementary Information.
Streamlines were used to visualize directional flow patterns; they visualize the course of single erythrocytes in the model (Fig. 2AC). In the clinically normal situation (Fig. 2A), the flow widened without turbulence from the arterial opening into the proximal IVS. In the pathological situations (Fig. 2B,C) - more pronounced in the IUGR simulation - the streamlines demonstrated the occurrence of a vortex, i.e., a recirculation zone close to the opening of the uterine spiral artery. In the clinically normal situation the number of streamlines in the arterial opening corresponded to 100% and, thus, 100% of the streamlines reached the top of the proximal IVS. In contrast, only 82.4% (IUGR) and 86.3% (IUGR/PE) of the streamlines reached the top of the proximal IVS in the pathological conditions, while 17.6% (IUGR) and 13.7% (IUGR/PE) of the streamlines shunted into the vortex.
(AF) show the situation in the models at the point of maximum systolic flow of a maternal heart cycle. (A and D) show the model of the clinically normal situation with a dilated arterial opening, (B and E) show the model of the situation in IUGR, and (C and F) show the model of the situation in IUGR/PE. In (AC) streamlines visualize the course of individual erythrocytes with the velocity color coded along the streamlines; the velocity color scale is shown in (C) (right of the proximal IVS). In (A) the streamlines run smoothly without turbulence at low speed from the arterial opening into the proximal IVS. In (B) and less pronounced in (C) the streamlines show the development of vortices in immediate proximity of the arterial opening. (DF) show color-coded iso-pressure surfaces, which are spaced by an interval of 0.05mm Hg; the color scale for (DF) (in mm Hg) is shown in (D) (left of the proximal IVS). In the clinically normal situation (D), the pressure decreases evenly through the proximal IVS toward the upper boundary (at the upper boundary, the pressure is defined as zero). In E and less pronounced in (F) the pressure remains elevated in a large part of the proximal IVS and then rapidly drops toward the upper boundary by compressing the iso-pressure surfaces at a short distance. The insert right of the proximal IVS in (F) visualizes the course of pressure along individual streamlines between the arterial opening and upper boundary for all three models.
Along a maternal heart cycle, the flow in the entry region into the proximal IVS of the simulated IUGR placenta sample remained laminar over a large time span and only became turbulent once a critical value of flow (Q=0.035ml/s) was exceeded. Below this threshold, the blood flow patterns were similar to those of the clinically normal situation. The threshold triggered appearance of the vortex in IUGR is shown in a movie in Supplementary Information (Movie S2).
The total pressure drop over the entire model, i.e., between the artery inlet and the reference point (pressure defined as p=0mm Hg) at the top of the proximal IVS was p=80mmHg in the clinically normal situation and 1.12 (IUGR) to 1.53 (IUGR/PE) times higher in the pathological cases (Table S2). This difference in pressure profiles could be subdivided into two general phenomena: (i) pressure drop along the spiralized arterial segments including the opening and (ii) pressure drop in the proximal IVS (see Figure S2 for definition).
The majority of the pressure drop occured along the spiralized artery and the arterial opening into the proximal IVS. The pressure drop in the arterial opening was 20 times steeper in IUGR and 14 times steeper in IUGR/PE compared to the clinically normal situation (Fig. 2DF, Table S1). The pressure drop in the arterial opening along a maternal heart cycle is presented in Supplementary Information (Movie S3a for the clinically normal situation and Movie S3b for IUGR).
At the entry into the proximal IVS, the pressure dropped to values below 1mm Hg. Iso-pressure surfaces were used to visualize the pressure in the proximal IVS; these surfaces connect points of the models with identical pressure. In the clinically normal situation, the iso-pressure surfaces were evenly distributed along the flow path through the proximal IVS and indicated a gradual and smooth decrease of pressure (Fig. 2D). Following the course of pressure along individual streamlines (Fig. 2, insert right of F), the pressure smoothly dropped throughout the proximal IVS in the clinically normal situation. In both pathological models, the pressure dropped biphasically, no longer gradually and continuously (Fig. 2E,F). In a substantial part of the proximal IVS close to the arterial opening, the pressure increased to high levels, which were elevated by a factor of 2.16 (IUGR) and 1.58 (IUGR/PE) compared to the clinically normal situation. Close to the upper boundary of the model, the pressure steeply dropped to reach the preset zero reference pressure at the upper model boundary, thereby compressing the iso-pressure surfaces at a short distance. This result indicates substantially higher pressure values at the middle and bottom of the proximal IVS compared to the clinically normal situation.
Doppler ultrasound from clinically normal, IUGR and IUGR/PE pregnancies were exemplary and typical waveforms for the investigated clinical syndromes (Fig. 3 and Table 1). The total blood flow delivered through both uterine arteries was in the range of 339593ml/min and in good agreement with previously reported estimates16,17. For all further calculations, the lower limit of 339ml/min was chosen to assume reasonable but minimally provocative conditions with respect to turbulence formation. The mean blood flow in a single spiral artery over one full maternal heart cycle equaled Q=0.031ml/s in the clinically normal situation, Q=0.027ml/s in IUGR and Q=0.017ml/s in IUGR/PE pregnancies (Table S2). The maximum flow value during maternal systole was similar in the three cases (Fig. 3). However, the flow was unstable and exhibited notching in the IUGR/PE case, which reduced the average blood flow per heart cycle. Consequently, the IUGR/PE case showed reduced vortex build-up as the critical flow value Q was only briefly exceeded during systole.
Time is shown on the x-axis, Doppler flow velocity in the uterine artery is shown on the left y-axis, and flow at the spiral artery inlet is shown on the right y-axis. Flow velocity in cm/s is the original output of the ultrasound analysis; flow velocity in ml/s is calculated from the original ultrasound data using the diameter of the respective arteria uterina. The entire waveforms were used to feed the dynamic flow model of the present study, and the interval of 1.2 to 2.0s (indicated by the black bar in the upper panel) was used for extraction of the evaluation data.
The dynamic flow models indicated areas of elevated wall shear stress at villous surfaces in the proximal IVS. To explore potentially elevated trophoblast shedding, we embedded a clinically normal and an IUGR placenta in paraffin in toto, divided the entire placenta block into four quarters and used one of these quarter placentas each for post-hoc histological analysis. The quarter placenta blocks were serially sectioned from the basal plate in a plane parallel to the chorionic plate. The serial sections were alternately stained either with hematoxylin and eosin (HE) or by immunohistochemical detection of cytokeratin 7 (CK7) as a trophoblast marker molecule (Fig. 4). In these sections, we could identify potential villous damage in IUGR, which appeared as cytokeratin-positive particles (partially as mononuclear particles) in the IVS but in a pronounced way in veins of intercotyledonary septa that drain the IVS.
Photomicrographs of histological sections of a clinically normal placenta (A,E,G,I,K) and an IUGR placenta (B,D,F,H,J,L) showing villi and veins (embedded in intercotyledonary septa) in a plane parallel to the chorionic plate (A,B). Villous sections of a clinically normal (A) and an IUGR (B) placenta demonstrate specific labeling of trophoblast by cytokeratin 7 (CK7). (CF) Tissue sections of veins (the vein walls are marked with black arrows) of a clinically normal (C,E) and an IUGR (D,F) placenta in low power overview. Sections shown in (C,D) were stained with hematoxylin-eosin (H,E), and sections shown in (E,F) were processed with immunohistochemistry using an anti-CK7 primary antibody. The black squares in (CF) mark the regions shown at higher magnification in (GJ). The square in (C) corresponds to (G) the one in (D) corresponds to (H) the one in (E) corresponds to (I) and the square in (F) corresponds to (J). (GJ) Cells inside the veins of a clinically normal (G,I) and an IUGR (H,J) placenta. (G,H) Cells inside the veins stained with HE and (I,J) CK7 positive cells are shown. Yellow arrow heads mark erythrocytes, and clear arrows mark isolated mononuclear cells. The black squares in (I and J) mark the regions shown at even higher magnification in (K and L). The square in I corresponds to (K) and the square in (J) corresponds to (L). (K,L) Cells inside the veins of a clinically normal (K) and an IUGR (L) placenta. Yellow arrowheads mark erythrocytes, clear red arrows mark isolated mononuclear cells and clear black arrows mark CK7 positive particles, putatively trophoblast shedding. The scale bar in (L) is 50m in (K,L), 100m in (A,B,GJ) and 800m in (CF).
The novel dynamic flow model of the present study extends far beyond the reach of the currently most advanced uteroplacental blood flow calculations12. It confirms the appearance of velocity jets inside the undilated arterial opening, but it substantially extends on aspects such as blood flow velocity and velocity jets in the IVS, directional flow (vortices), pressure analysis and wall shear stress. None of these aspects were previously simulated or determined. The morphological stage covered the spiralized artery and a substantial part of the post-arterial IVS and thus an assembly with the most relevant morphological features of the uteroplacental inflow region at the microscopic scale7. However, there are still limitations and specific assumptions which have to be kept in mind when interpreting the data of the present study. The model does not include fetoplacental vascularization or any aspects of the interplay of uteroplacental and fetoplacental circulations. It concentrates on a single uteroplacental inflow region, and the interplay with neighboring inflow regions was not considered. Although the height of 2mm of the proximal IVS above the spiral artery opening considered in the model is substantial at the microscopic scale, this is at best one quarter of the full distance between the chorionic and the basal plates. The boundaries inevitably resulting from the still limited model dimension can cause boundary effects. The zero pressure setting of the models at the upper model boundaries of the proximal IVS is, e.g., decreasing the absolute pressure values (but not the pressure relations between the simulations) in proximity of the boundary such that the absolute pressure levels in the proximal IVS might be substantially lower than those in the proximal IVS in vivo. The present study was undertaken as a pilot which provides quantitatively formulated hypotheses for later experimental validation under clinical or laboratory conditions. Since the models of the present study translate strict physical rules into a single placental setting, experimental validation has to focus on the variability due to various settings, but is beyond the scope of the present study.
Where reasonably possible, the basic assumptions of the present model, e.g., the dimensions and flow data of the uterine artery12, the number of uterine spiral arteries8,12,18 per placenta and parameters for blood and artery dilation, were kept consistent with previous reports12 to ensure and maintain the best possible backward comparability. The total flow data of our model were chosen at the lower limit of our flow calculations, though knowing that higher flow would mean higher cargo delivery to the placenta. Flow was kept intentionally low to set a frame of conservative flow assumptions in terms of the occurrence of turbulence. Keeping these considerations in mind, the present study outlines four areas in which the simulation of pathological conditions indicated differences from the clinically normal situation that need to be discussed (Figs 1 and 2): (i) formation of high-velocity jets at the arterial opening and their projection into the neighboring proximal IVS; (ii) higher wall shear stress and its possible consequences, particularly at surface areas of the proximal IVS where the high-velocity jets hit the villous tree; (iii) the formation of recirculation zones (vortices) in the entry region; and (iv) regions of the proximal IVS with elevated intervillous blood pressure.
The velocity jets observed in the pathological models of the present study developed inside the undilated opening of the spiral artery. The present model did not require assumptions on this opening because the morphologic stage was constructed from an undilated opening found in an IUGR placenta. Nevertheless, this real diameter corresponds well with the data used by others12. Our simulations confirm the Hagen-Poiseuille-based calculations of ref. 12 but are able to extend the simulation into the proximal IVS. The hypothesis of ref. 12 that the intravascular velocity jets extend substantially into the IVS could be confirmed by our extended simulations. All these findings together support our idea that dilation of the final segments of spiral arteries in human placentas can be considered to be a way to avoid or minimize the occurrence of velocity jets. This could be functionally essential because high-velocity regions such as the observed jets are contradictory to slowly progressing flow patterns known from other diffusion-governed organs with high metabolic activity, e.g., the liver19,20,21. Our models also showed that the occurrence of velocity jets was restricted to the peak systolic phases of the maternal heart cycle. This result indicates that not only the morphology of the arterial opening but also preuterine circulatory conditions could influence the occurrence of velocity jets.
In addition, the fluid dynamic model of the present study revealed increased wall shear stress at the villous surface in the inflow regions of the placenta (Fig. 1DF; Movie S1a,b), particularly in regions that were reached by high-velocity jets. High wall shear stress indicates that some regions of the villous tree could become critically stressed. Likely, this has consequences for the sensitive syncytial trophoblast surface, which covers the villous tree and enables gas and nutrient exchange during pregnancy. Particularly in the vicinity of the velocity jets in IUGR and to a minor extent in IUGR/PE (Fig. 2), high wall shear stress could damage this epithelium or at least challenge its cellular turnover22,23. Effects of wall shear stress on trophoblast could potentially be mediated by certain mechanotransduction pathways that were recently identified in trophoblast23.
Post-hoc histological examination of uterine veins in intercotyledonary septa revealed evidence of increased damage at the trophoblast surface (Fig. 4). Increased amounts of cytokeratin-positive particles, partially also with a mononuclear appearance, were found in the intercotyledonary veins of a placenta from a patient with IUGR. The mixture of anuclear and seemingly mononuclear particles is not atypical for debris of syncytiotrophoblast. Such phenomena (including the appearance of mononuclear cytokeratin-positive particles) are known from protocols for isolating primary trophoblasts24. Because larger particles such as those observed here would stick in lung capillaries and be degraded there, elevated levels of (free) fetal DNA in the maternal circulation would be a possible endpoint of trophoblast shedding. This has indeed been reported as being associated with IUGR and IUGR/PE25,26,27,28.
Corresponding to the pressure drop at the arterial opening and the velocity jet in the center axis of the undilated spiral artery openings, vortices appeared in the pathological models, in direct proximity of the arterial opening. Such vortices were not described by other models and principally cannot be discovered using the limiting assumptions of laminar flow only by the Hagen-Poiseuille equation. In a vortex, the blood is circulating rather than slowly progressing along a surface. According to the data of the model of the present study, approximately 13.717.6% of the blood entering the placental IVS during the systolic peak phase of the maternal heart cycle could shunt into a vortex. Specifically, vortices can capture a substantial part of the inflow volume into cycling dead volumes, which would reach the feto-maternal exchange zone at the villous surface with delay and could possibly contribute to reduced, e.g., oxygen transfer6,22,29.
The simulations of the present study provided evidence that areas with higher than clinically normal blood pressure might extend deep into the IVS in vivo. However, the absolute values observed in the models in the proximal IVS should be interpreted with caution. They were influenced by boundary effects and the dimensions of the model and likely underestimated the true pressures in the IVS in vivo substantially and systematically. The pressures in the model of the present study would come closer to the in vivo values if (i) the proximal IVS would have substantially larger dimensions, ideally composing the entire villous zone between the basal plate and the chorionic plate, and (ii) the venous pressure in the draining veins would replace the zero pressure setting.
Although the blood pressure in the IVS in vivo might be higher than in the corresponding models, the difference in the outcome between the simulations of clinically normal and pathological conditions indicated a profound effect of the arterial opening on the pressure in the IVS. While pressure dropped more steeply at the arterial opening in the pathological cases than in the clinically normal situation, the end pressure in the proximal IVS of the IUGR and IUGR/PE conditions was higher than in the proximal IVS of the clinically normal situation. Increased blood pressure in the IVS, as indicated for the first time in the models of the present study, is unexpected in current pathogenetic concepts and could have a substantial impact on villous topology30 and placental function. These circumstances can lead to blood retention in the IVS next to the opening of the uterine spiral arteries. Because blood is not compressible, a large blood lake arising next to the opening of the uterine spiral arteries could operate as a blood pressure cuff in the IVS or could lead to a dead volume, which corresponds to the vortex formation outlined above.
In vivo, zones with blood pressure elevated above normal in the IVS are potentially extending far deeper in the placental tissue than was revealed by the dimensions of the fluid dynamics model of the present study (Fig. 1AC). Elevated blood pressure in the IVS could potentially be an additional and independent factor compromising placental perfusion. The capillaries in the peripheral parts of the villous tree are directly exposed to the intervillous pressure and are kept open for fetal perfusion only by their internal pressure provided by the fetal heart.
These capillaries have no determinant of their diameter other than the relation of their capillary blood pressure to the blood pressure in the IVS31. It can be expected that the differences between the blood pressure in the IVS and the blood pressure in the fetal capillaries are discrete. Elevated blood pressure in the IVS, as suggested by the models of the present study, could thus be able to reduce fetal capillary diameters and increase fetal vascular resistance with the full 4th degree exponential power of diameter changes in this context32. Pressure peaks or constant pressure above a critical value in the IVS would tend (i) to compress these capillaries, preferably during the low-pressure phases of the fetal heart cycle, (ii) to substantially compromise fetal blood flow, and (iii) could possibly be the cause of umbilical reverse Doppler flow in the third trimester of gestation in pregnancies affected by severe IUGR.
Clinical Doppler umbilical flow parameters such as end diastolic flow and pulsatility index are clinical signals of micro-circulatory resistance properties inside the fetal vascular bed of the villous tree and are important for managing apparent disease to determine an optimal time of delivery33. Absent or reversed end diastolic umbilical flow could be interpreted in this context as a condition in which, e.g., an elevated intervillous pressure compresses intravillous capillaries and contributes to critical reduction of fetal blood flow and could even be a factor helping to shift arterial volume back at the end of the fetal diastole4,33,34.
The dynamic flow model implemented in the present study indicates that insufficient remodeling of uterine spiral arteries could have a multitude of effects on the blood flow in spiral arteries and the IVS. The villous surface appears to be exposed to increased mechanical stress by velocity jets and elevated wall shear stress, which correspond to intensified trophoblast shedding and elevated levels of fetal DNA in maternal blood. The present study underlines this concept by histopathologically demonstrated increased trophoblast shedding. Potentially even more important are those modifications of the blood flow in the IVS which have the potential to directly impair the placental exchange functions. These are (i) turbulent flow conditions, particularly vortices at the arterial openings, which are an equivalent of functional dead volumes with suboptimal exchange capacity, and (ii) elevated blood pressure in the IVS, which could potentially become a risk for fetoplacental perfusion as soon as fetal capillaries become compressed or even collapse. In view of the concept of ischemic placental disease, the present study demonstrates possible pathogenetic mechanisms that are capable of linking the early origins (insufficient arterial remodeling) to the considerably later development of clinically symptomatic placental disease1,35,36.
All clinical and morphological examinations and tissue collections were part of a larger study approved by the ethics committee of the Ludwig-Maximilians-University of Munich (LMU Munich, Germany) under number 47812. The placentas were collected at the Department of Obstetrics and Gynecology of the hospital Dritter Orden, Munich, Germany. The placentas were allocated to clinical groups by the obstetricians based on clinical information regarding the pregnancy and delivery. The collection included clinically normal placentas and placentas of pregnancies with intrauterine growth retardation (IUGR) with absence or presence of preeclampsia (PE). An IUGR was diagnosed if the growth data of the fetus determined by ultrasound (e.g., body length and head circumference) were above the 10th growth percentile during the first two trimesters and then dropped below the 10th growth percentile4,11. In addition to IUGR, PE was defined by arterial hypertension (140/90mmHg) with onset beyond the 20th week of gestation or substantial proteinuria (300mg/24h) after the 20th week of gestation4,11. The study protocol allowed for the inclusion of clinical data such as Doppler data if anonymity is maintained. Placentas were collected after informed consent of mothers/parents was obtained. Placentas were excluded when no informed consent of the mothers/parents could be obtained, when the language skills of the mothers/parents limited understanding of information regarding the study, or when psychiatric problems or any other condition caused doubts regarding the ability of the mothers/parents to independently decide. All work was conducted according to relevant guidelines and regulations, and all data were anonymized. Additional details are outlined in Supplementary Information (Sources of tissue and clinical data; ethical approval).
During routine histopathological examination, a long segment of a spiral artery and the adjacent region of the villous tree and intervillous space were located in a placenta of a patient with IUGR (Table S1). Serial sectioning and reconstruction were used to three-dimensionally reconstruct the spiral artery and the inflow region proximal to the arterial opening (proximal IVS). Technical details of this procedure are outlined in Supplementary Information (Generation of the 3D tissue reconstruction).
For post-hoc histological analysis of potential tissue damage in specific areas of the villous tree, one additional placenta from a clinically normal pregnancy and one additional placenta of a patient with IUGR (Table S1) were used. Both placentas were fixed and embedded in paraffin, each in toto. Serial sections, of one quarter of each placenta, were taken in a plane that was parallel to the chorionic plate, beginning at the basal plate and stained alternately with hematoxylin-eosin or used for immunohistochemical detection of the trophoblast marker molecule cytokeratin 77. Technical details of this procedure are outlined in Supplementary Information (Tissue processing for post-hoc histological analysis; Immunohistochemistry protocol). The Doppler data of uterine arteries used in the dynamic flow model of the present study were collected from three patients with either uncomplicated pregnancy, IUGR or IUGR/PE, respectively (Doppler data and clinical data of these patients are shown in Fig. 3).
Blood flow in the left (l) and right (r) uterine artery was calculated using the velocity coming from Doppler ultrasound measurements in the left and right uterine artery and a cross sectional area AAU with a diameter of dAU=3.0mm18,37 as
We used three typical waveforms from clinically normal, IUGR and IUGR/PE pregnancies (collected by Tanja Ruebelmann, Dritter Orden hospital). Assuming that this flow is distributed equally to n=130 uterine spiral arteries in the placenta38, blood flow through a single uterine spiral artery reads as
represents the deformation rate tensor of the fluid. These equations are solved via stabilized finite elements39,40 in our in-house code, which has been used successfully in several biomedical flow applications41,42.
In this study, blood was modeled as a Newtonian fluid with a kinematic viscosity of 6 mPas, which corresponds to the upper viscous limit in12 and a density of =1.055106g/mm3. At the artery inlet, velocity waveforms extracted from Doppler ultrasound measurements at the left AU were prescribed (see previous section). At the walls of the artery and the bottom of the placenta, no-slip boundary conditions were used, and at the sides of the intervillous space, slip boundary conditions were used. At the top surface of the modeled IVS, a zero-traction boundary condition was applied. All computations were performed with a time step of t=104s and a convergence criterion of 105 in the nonlinear residual. The simulations were performed on the SuperMuc Petascale System (Project pr83te: High Performance Methods for Computational Fluid Dynamics including Multiphysics Scenarios) of the Leibniz Supercomputing Center (Garching, Germany) and were analyzed and visualized using Paraview software43,44 in the morphological stage of the present study, which is in agreemement with general anatomic data45,46,47,48,49,50 (for details see Supplementary Information).
How to cite this article: Roth, C. J. et al. Dynamic modeling of uteroplacental blood flow in IUGR indicates vortices and elevated pressure in the intervillous space a pilot study. Sci. Rep. 7, 40771; doi: 10.1038/srep40771 (2017).
Ananth, C. V. Ischemic placental disease: A unifying concept for preeclampsia, intrauterine growth restriction, and placental abruption. Ischemic placental disease. Semin Perinatol. 38, 131132 (2014).
Burton, G. J., Woods, A. W., Jauniaux, E. & Kingdom, J. C. P. Rheological and Physiological Consequences of Conversion of the Maternal Spiral Arteries for Blood Flow during Human Pregnancy. Placenta 30, 473482 (2009).
Pennati, G., Socci, L., Rigano, S., Boito, S. & Ferrazzi, E. Computational Patient-Specific Models Based on 3-D Ultrasound Data to Quantify Uterine Arterial Flow During Pregnancy. IEEE Transactions on Medical Imaging 27, 17151722 (2008).
Ferrazzi, E., Rigano, S., Padoan, A., Boito, S., Pennati, G. & Galan, H. L. Uterine Artery Blood Flow Volume in Pregnant Women with an Abnormal Pulsatility Index of the Uterine Arteries Delivering Normal or Intrauterine Growth Restricted Newborns. Placenta 32, 487492 (2011).
Rigano, S., Ferrazzi, E., Boito, S., Pennati, G., Padoan, A. & Galan, H. Blood Flow Volume of Uterine Arteries in Human Pregnancies Determined Using 3D and Bi-Dimensional Imaging, Angio-Doppler, and Fluid-Dynamic Modeling. Placenta 31, 3743 (2010).
Konje, J. C., Kaufmann, P., Bell, S. C. & Taylor, D. J. A longitudinal study of quantitative uterine blood flow with the use of color power angiography in appropriate for gestational age pregnancies. Am J Obstet Gynecol 185, 608613 (2011).
Munk, O. L., Bass, L., Feng, H. & Keiding, S. Determination of regional flow by use of intravascular PET tracers: microvascular theory and experimental validation for pig livers. J Nucl Med. 44, 18621870 (2003).
Haeussner, E., Schmitz, C., Frank, H. G. & Edler von Koch, F. Novel 3D light microscopic analysis of IUGR placentas points to a morphological correlate of compensated ischemic placental disease in humans. Sci Rep. 6, 24004 (2016).
Gomez, O. et al. Sequential changes in uterine artery blood flow pattern between the first and second trimesters of gestation in relation to pregnancy outcome. Ultrasound Obstet Gynecol. 28, 802808 (2006).
Palmer, S. K., Zamudio, S., Coffin, C., Parker, S., Stamm, E. & Moore, L. G. Quantitative estimation of human uterine artery blood flow and pelvic blood flow from redistribution in pregnancy. Obstet Gynecol. 80, 10001006 (1992).
Gravemeier, V., Gee, M. W., Kronbichler, M. & Wall, W. A. An algebraic variational multiscale-multigrid method for large eddy simulation of turbulent flow. Comput Methods Appl Mech Engrg. 199, 853864 (2010).
Hughes, T. J. R., Scovazzi, G. & Franca, L. P. Multiscale and stabilized methods. In Stein, E., de Borst, R., Hughes, T. J. R. editors. Encyclopedia of Computational Mechanics. pp. 559 (John Wiley & Sons, Chichester, 2004).
Comerford, A., Gravemeier, V. & Wall, W. A. An algebraic variational multiscale- multigrid method for large-eddy simulation of turbulent pulsatile flows in complex geometries with detailed insight into pulmonary airway flow. Int J Numer Meth Fluids. 71, 12071225 (2012).
Geuzaine, C. & Remacle, J. F. Gmsh: a three-dimensional finite element mesh generator with built-in pre- and post-processing facilities. Int J Numer Methods Eng. 79, 13091331 Link: http://gmsh.info/, Accessed: 14-03-2016 (2009).
The authors acknowledge the skillful technical assistance and diligent work of the entire team of technicians of the Department of Anatomy II at LMU Munich, namely, B. Aschauer, A. Baltruschat, S. Kerling, B. Mosler and S. Tost. We would also like to express our thanks to all donators and the obstetricians, midwives and nurses of the Dritter Orden hospital (Munich, Germany) who enabled the clinical work of this study with great care and engagement.
C.R. and E.H. performed the analysis. E.H., C.R., W.W. and H.-G.F. designed the study. F.v.K. and T.R. were responsible for obstetrical tasks. C.R., E.H. and H.-G.F. prepared figures and tables. E.H., C.R., C.S., F.v.K., W.W. and H.-G.F. wrote the manuscript.
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articles Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
Roth, C., Haeussner, E., Ruebelmann, T. et al. Dynamic modeling of uteroplacental blood flow in IUGR indicates vortices and elevated pressure in the intervillous space a pilot study. Sci Rep 7, 40771 (2017). https://doi.org/10.1038/srep40771