Atherosis of Trophoblast Type: A Specific Form of Decidual Vasculopathy Distinct From Atherosis of Macrophage Type

The study is exempt from institutional review board approval according to section 46.101(b) of 45CFR 46, which states that research involving the study of existing pathologic and diagnostic specimens in such a manner that subjects cannot be identified is exempt from the Department of Health and Human Services Protection of Human Research Subjects. Placental tissues submitted for pathology examination for a variety of clinical indications in 2021 were included in the study using the routine hematoxylin-eosin (H&E) stain. Paraffin-embedded tissues from routine surgical pathology specimens and routine H&E-stained pathology slides were examined by light microscopy, using the Amsterdam criteria for placental examination.^1,2  Four blocks of placenta, umbilical cord, and membranes are routinely submitted including 1 block of membrane roll with 2 sections of umbilical cord, 2 blocks of placental parenchyma including fetal surface and maternal surface, and 1 block of horizontal section of decidua surface (en face), using the Khong method.¹⁷  Immunohistochemical staining for CD56 and CD68 was used to highlight the specific cell types when needed. Immunohistochemical staining procedures were performed on paraffin-embedded tissues with the Leica Biosystems Bond III automated immunostaining system following the manufacturer's instructions.¹¹  CD56 and CD68 monoclonal antibodies were purchased for clinical applications from Dako Agilent under the catalogue Nos. M730401-2 and M087601-2.

A total of 1322 placentas with corresponding neonatal data from January 2021 to August 2021 were collected from the medical records, including 96 normal pregnancies. The placental pathology data were entered into an Excel spreadsheet (Microsoft Corporation) at the time of pathology examination, and the neonatal sex, birth weight, birth length, and head circumference were subsequently retrieved from the medical records. Twin pregnancies with identical sex were included in the study, and the twin pregnancies with different sex were excluded.

Separately, a database of preeclampsia/pregnancy-induced hypertension was established between 2017 and 2020, consisting of placental pathologic features and clinical features. The data include 619 placentas of late-onset preeclampsia and 133 normal controls. A portion of these data were previously published.¹¹  Atherosis within this data set was collectively named as classic type without further separation of atherosis of macrophage type and atherosis of trophoblast type. Early-onset preeclampsia was excluded from the current study owing to the confounding factor of gestational age affecting the placental and neonatal weight.

Statistical analysis was performed by using various programs of the R-package including baseline characteristic table and multivariant analysis of variance tests (http://statistics4everyone.blogspot.com/2018/01/fathers-data-visualization.html; accessed July 28, 2021).

In typical acute atherosis, foamy macrophages can be identified within the arterial walls, and these macrophages can be highlighted by immunohistochemical staining for CD68, a macrophage marker (Figure 1, A and B). Acute atherosis was initially described by Zeek and Assali,⁶  and the terminology remains essentially unchanged to date with little controversy. In a subset of acute atherosis, the foamy cells were not proven to be macrophages but rather endovascular trophoblasts with unique CD56 expression (Figure 1, C and D).¹¹  CD56 expression is restricted on the endovascular trophoblasts only, in addition to decidual natural killer cells, thus separating the population of extravillous trophoblasts from that of endovascular trophoblasts.¹¹  Atherosis of macrophage type shares significant morphologic similarities to atherosis of trophoblast type, and both can show foamy cell changes in the background of eosinophilic fibrinoid arterial media (fibrinoid medial necrosis). The morphologic differences can be appreciated with slightly basophilic cytoplasm of intramural or endovascular trophoblasts under light microscopy, whereas the clear foamy cytoplasm of macrophages is normally identified. In their initial description of atherosis of macrophage type, Zeek and Assali⁶  emphasized the anatomic location of membrane rolls, consistent with the lack of trophoblast invasion and trophoblast-directed spiral artery remodeling in the decidual capsularis.

Based on the separation of atherosis of macrophage type from trophoblast type, decidual vasculopathy can be classified as classic type, mixed type, and mural arterial hypertrophy as in Table 1. Atherosis and fibrinoid medial necrosis coexist commonly, and pure atherosis without fibrinoid medial necrosis is rare. Both atherosis and fibrinoid medial necrosis are historic terms, and so named as “classic type.” The term atherosis of trophoblast type still applies even when the background of fibrinoid medial necrosis in late pregnancy is absent, because the pathogenic mechanism is the persistence of vascular trophoblasts in third trimester. Mural arterial hypertrophy is a separate entity that is independent of trophoblast invasion, and it typically occurs in the small-caliber arteries in either decidua basalis or capsularis. Mural arterial hypertrophy may relate to hormonal influence (progesterone), as similar changes of small arteries can be seen in secretory endometrium without pregnancy.¹⁸  It is worth noting that morphologic changes of mural hypertrophy in superficial myometrium were used as evidence of “failed trophoblast invasion” in preeclampsia.^15,16 

Mixed-type vasculopathy is defined as the presence of both atherosis and mural arterial hypertrophy in the same or different anatomic locations such as decidual basalis and/or decidual capsularis. The morphologic features of atherosis and mural arterial hypertrophy are different with different implications for pathogenesis, and the combination of both morphologic features (Figure 2, A through F, and Figure 3, A through D) in the same case represents a distinct risk for both mother and fetus. In our data, mixed-type vasculopathy was associated with the most significant changes in placental weight, neonatal weight, and growth, as well as placental pathology (later section).

This new classification for decidual vasculopathy incorporated new evidence of pathogenesis of maternal vascular transformation from early to late pregnancy and provided new insight in understanding the interaction between the trophoblasts, endothelial cells, and smooth muscle cells and the potential risk of maternal cardiovascular disease in later life after pregnancy.

Practically, there are morphologic variants of atherosis of trophoblast type. The examples of these vascular changes are shown in Figure 2. In membrane rolls the decidual vessels are not remodeled by endovascular trophoblasts, and the vascular changes, including hypertrophic/hyperplastic and/or atherosis/fibrinoid medial necrosis, are not directly related to trophoblasts; therefore, atherosis of macrophage type and mural hypertrophy/hyperplasia are present exclusively. In decidua basalis where the extensive spiral artery remodeling occurred at early implantation, 3 types of vascular changes—both atherosis of trophoblasts and atherosis of macrophages, as well as mural hypertrophy—can be identified. In our experience, atherosis of trophoblast type is far more common than that of macrophage type, suggesting significantly different pathogenic mechanisms. Unlike atherosis of macrophage type, atherosis of trophoblast type does not seem to play significant roles in the pathogenesis of preeclampsia or hypertensive disorders of pregnancy.

A total of 1322 placentas with clinical, pathologic, and neonatal data were collected in 2021 chronologically in light of new classification of decidual vasculopathy (Table 2). In all, 468 of 1322 placentas shows features of various decidual vasculopathies. Atherosis of macrophage type (P = .04), mixed-type vasculopathy (P = .001), and mural hypertrophy (P = .001) showed statistically significant association with preeclampsia/pregnancy-induced hypertension (PIH). All types of decidual vasculopathy were associated with decreased placental weight, neonatal weight, and intrauterine growth restriction (P = .001). Most of the statistically significant associations between the decidual vasculopathy and clinical and placental pathologic features were previously known, and our current data are consistent with previous knowledge. There was statistically significant association of various types of decidual vasculopathy with body mass index. There were statistically significant associations between the specific vasculopathy type and placental weight at delivery. It is worth noting that mixed-type vasculopathy is the most significant placental lesion and it is associated with the lowest placental weight, neonatal weight, neonatal birth length, and most significant placental pathologies.

Atherosis of trophoblast type with persistence of vascular trophoblasts in late pregnancy leads to questions of failed trophoblast cell death, or failed endothelial cell/smooth muscle cell regeneration or both. We explored the potential association between the neonatal sex and atherosis of trophoblast type, as vascular trophoblasts are of fetal origin in maternal circulation. Trophoblast invasion into spiral artery and persistence of endovascular trophoblasts within maternal spiral artery may relate to sex dimorphism. As shown in Table 3, there were 651 female neonates and 671 male neonates in the current study data. There is a statistically significant association between female neonates and mural hypertrophy (P = .001). There is no statistically significant difference in frequencies of other types of decidual vasculopathy between male and female neonates. There is no difference in placental pathology and laboratory test results except for neonatal birth weight, birth length, and birth head circumference between male and female sex. There is a statistically higher incidence of cesarean delivery for male neonates than for female neonates.

Atherosis of macrophage type is known to be associated with preeclampsia/PIH.⁶  We examined our current dataset to see if there is an association of atherosis of trophoblast type with preeclampsia/PIH (Table 4). There is a statistically significant association between atherosis of macrophage type, mixed-type vasculopathy, mural hypertrophy, and preeclampsia (P = .04, P = .001, and P = .001, respectively). There is no association between atherosis of trophoblast type and preeclampsia (P = .22). There are many positive and negative associations between clinical and pathologic features with preeclampsia, and most of these associations are previously known (Table 4). It is important to note that the case number of atherosis of macrophage type is small in current data (n = 19), and the statistical association with preeclampsia/PIH should be interpreted with caution.

Separately, we have collected 619 placentas of late-onset preeclampsia and 133 normal controls to determine the placental weight and associated various type of decidual vasculopathy.¹¹  The data grouped the classic-type vasculopathy as 1 entity and did not separate atherosis of macrophage type from atherosis of trophoblast type (Figure 4). There is a statistically significant association between classic-type vasculopathy and the placental weight. Classic-type vasculopathy is associated with lower placental weight in comparison to normal healthy pregnant controls. Mixed-type vasculopathy is statistically associated with the lowest placental weight.

Acute atherosis was first described by Zeek and Assali⁶  in 1950, and there is very little debate about the diagnostic criteria or clinical significance of Zeek atherosis. Acute atherosis has been associated with preeclampsia or hypertensive disorders of pregnancy, and numerous subsequent studies have confirmed this association, although the pathogenesis is unclear.^6,9,10,19,20  Oxidative stress of placental bed vessels in preeclampsia has been extensively studied owing to the similarity of acute atherosis with adult atherosclerosis, and it is clear that oxidative stress plays important roles in formation of atherosis (macrophage migration and deposit within the arterial wall).^8,21–23  It is also clear that lipid plays a role in formation of foamy macrophages in atherosis (oil Red-O), similar to atherosclerosis.^24–26 

Atherosis of trophoblast type is so named because that in a subset of acute atherosis previously described, the intramural or endovascular foamy cells are of trophoblast origin, rather than macrophage origin.^11,20,27  Identification of the trophoblast origin of some of the foamy cells leads to a different set of questions of pathogenesis. Endovascular or intramural trophoblasts have been described many decades ago, and identification of these trophoblasts under light microscopy presents no problem for pathologists.⁵  The critical link between the endovascular or intramural trophoblasts to foamy cells and atherosis is that vascular trophoblasts express CD56 exclusively, in addition to conventional trophoblast markers such as cytokeratins, p63, and HLA-G, and these vascular trophoblasts originate from the “plugging cells” of the spiral artery in early pregnancy.^11,28  Spiral artery remodeling is an essential physiological step of normal pregnancy. Naturally, the presence of vascular trophoblasts in late pregnancy has been used as evidence of “remodeled spiral artery” in contrast to failed spiral artery remodeling. Remodeled spiral artery in late pregnancy has been called “physiologic change.”^9,15,16,20  However, plugging cells from early implantation are known to disappear starting at approximately 7 weeks, and the uteroplacental circulation is fully established at the end of the first trimester.²⁸  Vascular trophoblasts are rarely seen in the third trimester of normal pregnancy.^5,11  Temporal and spatial cellular functions of vascular trophoblasts have not been taken into consideration in the historic studies.^15,29  Persistence of vascular trophoblasts in late pregnancy (third trimester) therefore represents abnormal (pathologic) changes in placenta rather than normal physiologic features. However, persistence of vascular trophoblasts in late pregnancy has not been systemically studied to date to our knowledge.

For a pathologist, the diagnostic criterion is critical for accurately categorizing the placental abnormalities. The changes in atherosis of trophoblast type should be similar to those of atherosis of macrophage type—that is, foamy cells in the background of fibrinoid medial necrosis of the arterial wall. In some cases, vascular trophoblasts are present without fibrinoid medial necrosis. These morphologic changes should also be called atherosis of trophoblast type, given the pathogenic mechanism remains as persistence of vascular trophoblasts in late pregnancy.

As atherosis of trophoblast type represents persistence of vascular trophoblasts in maternal circulation, we have attempted to examine the role of sex dimorphism in our data. Our current data show no evidence of sex difference in frequencies of atherosis of trophoblast type. However, study of sex dimorphism requires a large number of cases to achieve meaningful difference.^30–32  Our current dataset may not contain enough cases, and the data collection is still ongoing with a potential update once we have adequate case numbers.

Our current data indicate that atherosis of trophoblast type is an abnormal finding in placental examination and should be stated as such in a pathology report. Persistence of vascular trophoblasts invariably affects the vascular involution after delivery, and it has been associated with postpartum hemorrhage.³³  It is unclear how long these vascular trophoblasts persist in the maternal arterial wall after delivery/pregnancy, and the presence of these vascular trophoblasts within the maternal spiral artery wall will likely affect the vascular stiffness and hemodynamics of maternal circulation after pregnancy.^34–36  The constellation of these maternal vascular changes is likely a significant risk factor for cardiovascular disease in later life after pregnancy. Further evidence is required to better assess the potential link of these morphologic changes of placental examination to better health of women after pregnancy.

Atherosis of trophoblast type represents a spectrum of morphologic changes consistent with abnormal placental pathology, and it is associated with various maternal pregnancy complications with distinct pathogenic mechanism related to trophoblast cell death and vascular regeneration.

This work is dedicated to the memory of Kurt Benirschke, MD, whose mentorship and encouragement in placental pathology has had a profound influence on the authors.