Histopathology in the placentae of women with antiphospholipid antibodies: A systematic review of the literature

Abstract

Background

Antiphospholipid antibodies (aPLs) are a heterogenous group of autoantibodies associated with recurrent miscarriage, stillbirth, foetal growth restriction and premature birth. The cause of obstetric morbidity in women with aPLs is not completely understood. Workers have attempted to understand the role of aPLs in obstetric morbidity by investigating the histopathology of placentae from aPL-positive women. However, it is unclear from these diverse, and at times contradictory reports what histopathological lesions are common in the placentae of women with aPLs. This systematic review was undertaken to generate a complete picture of the placental features associated with aPLs in an attempt to understand the pathological processes that occur in pregnancies affected by aPLs.

Methods

Pubmed, Scopus, Web of Science and Embase were searched on the 27th November 2014 using the keywords “placenta” OR “trophoblast” AND “antiphospholipid antibody” OR “antiphospholipid antibody syndrome”. Records that were relevant and eligible were qualitatively assessed and given a score out of 24.

Results

Of the 1112 records that were retrieved from the systematic search, 34 records were eligible for review, and were qualitatively scored. Of the 44 histopathological features that were reported in 580 placentae from aPL-positive women, six features appeared to be more common in the placentae of aPL-positive women compared to control women, including: placental infarction, impaired spiral artery remodelling, decidual inflammation, increased syncytial knots, decreased vasculosyncytial membranes and the deposition of complement split product C4d.

Conclusion

Based on the evidence in this systematic review, a human placental aPL fingerprint has been proposed. The diversity of the human placental aPL fingerprint suggests that multiple pathological processes may occur in pregnancies affected by aPL.

Introduction

Antiphospholipid antibodies (aPLs) are a heterogeneous group of autoantibodies that are found in 0–5% of the general population [1], [2] and in up to 40% of women suffering from stillbirth and/or recurrent miscarriage [3], [4], [5], [6], [7]. Antiphospholipid antibodies are also significantly associated with preeclampsia, intrauterine growth restriction and premature birth [4], [7], [8], [9], [10] as well as arterial and venous thrombosis [11], [12], [13], [14]. Antiphospholipid antibodies react with a complex antigen involving anionic phospholipids and phospholipid-binding proteins, such as β₂ glycoprotein I. More than 18 aPLs have been reported, although only anti-β₂ glycoprotein I antibodies, anticardiolipin antibodies, and the lupus anticoagulant are considered clinically relevant in most individuals. Women who experience arterial/venous thrombosis and/or obstetric morbidity and test positive for one or more of the clinically relevant aPLs are defined as having the antiphospholipid antibody syndrome (APS) [15].

Obstetric morbidity in women with the APS may be due to the detrimental effects of aPLs on the placenta, an organ whose function is critical to the success of pregnancy. The human placenta consists of an extensive network of finger-like projections, or placental villi, most of which float in the maternal blood-filled intervillous space. All floating placental villi are covered by a multinucleated cell called the syncytiotrophoblast. The syncytiotrophoblast is a sessile, terminally differentiated epithelium that is maintained by the continuous incorporation of nuclei and cellular machinery from underlying villous cytotrophoblasts. Villous cytotrophoblasts proliferate and then fuse with the syncytiotrophoblast in a process called syncytialisation. Three to four weeks following syncytialisation [16], syncytiotrophoblast nuclei cluster, become increasingly heterochromatic and undergo DNA fragmentation [17], [18]. Histologically, clusters of heterochromatic syncytiotrophoblast nuclei are called syncytial knots [19]. Syncytial knots and smaller fragments of the syncytiotrophoblast (collectively termed trophoblast debris) may be extruded from the surface of the syncytiotrophoblast into the maternal blood [20], [21] in a process akin to the apoptotic blebbing of mononuclear cells [22]. Beneath the syncytiotrophoblast and villous cytotrophoblasts is the stromal core of the placental villus, which contains foetal blood vessels, fibroblasts and macrophages.

In some placental villi, specialised populations of villous cytotrophoblasts give rise to extravillous trophoblasts, which penetrate through the syncytiotrophoblast and grow into the maternal decidua, thereby anchoring the placenta to the uterus [19]. Extravillous trophoblasts also invade the uterine spiral arteries, initially forming plugs that partially or fully occlude the lumens of spiral arteries [23], and later transforming the spiral arteries into large-bore, non-vasoactive vessels that maximize maternal blood supply to the placenta [19].

In vitro work has demonstrated that aPLs adversely affect trophoblast function [24]. Specifically, treatment of trophoblasts or placental villous explants with aPLs has been shown to decrease trophoblast proliferation [25], syncytialisation [26], [27], and invasion [27], [28], [29], [30], [31], [32], [33]. Antiphospholipid antibodies also appear to affect trophoblast death [24], as treatment of placental explants with aPLs increases trophoblast apoptosis [34], [35], [36] and the release of trophoblast debris from the syncytiotrophoblast [37].

In addition to affecting trophoblast function and death, aPLs may also contribute to placental dysfunction and thus, obstetric morbidity by triggering thrombosis in the intervillous space. Placental thrombosis is conceivable given that aPLs promote thrombosis in the systemic circulation [11], [12], [13], [14], and since the intervillous space is a site of blood turbulence and blood stasis. Moreover, the syncytiotrophoblast naturally exteriorises phosphatidylserine [38], [39], [40], a trigger of contact-dependent coagulation [41]. In normal pregnancies, exteriorised phosphatidylserine may be protected from maternal coagulation factors by annexin V, a phosphatidylserine-binding protein [42] that is thought to form a lattice-like shield over the surface of the syncytiotrophoblast [43]. The annexin V shield may be disrupted in pregnancies affected by aPLs, as suggested by experiments performed in vitro, which have demonstrated that aPLs displace annexin V from the surface of placental explants [44], [45] or syncytialised trophoblasts [46], [47], [48], [49]. Following disruption of the annexin V shield, syncytialised trophoblasts bound more prothrombin [48] and more rapidly formed clots in cell-surface plasma coagulation assays [46], [47], suggesting that the changes in trophoblastic annexin V had pro-thrombotic consequences in vitro. Excessive coagulation on the surface of the syncytiotrophoblast may manifest as the deposition of fibrin around placental villi (perivillous fibrin deposition), or the formation of thrombi in the intervillous space. Both lesions have the capacity to obliterate the intervillous space, and may therefore render placental villous tissue ischaemic and functionally void [23].

Placental dysfunction and foetal demise in pregnancies complicated by aPLs may also reflect excessive inflammation within the uteroplacental site. The inflammatory insult may be driven by the activation of complement, as demonstrated by a series of elegant studies where pregnant mice were passively immunised with IgG from individuals with the APS [50]. Following the passive immunisation of pregnant mice, aPLs are thought to deposit within the uteroplacental site where they activate the classical complement cascade [51]. Activation of the classical complement cascade results in cleavage of complement component C3 into C3b, which is deposited on the surface of decidual cells [50]. Membrane-bound C3b then cleaves complement component C5 into C5a, an anaphylatoxin that promotes neutrophil activation and migration [51], that increases the expression of tissue factor and leads to neutrophil oxidative burst [52]. The reactive oxygen species, proteolytic enzymes and other inflammatory mediators released by neutrophils inflict damage within the uteroplacental site and lead to foetal demise in this mouse model of human APS pregnancy.

Antiphospholipid antibodies may also elicit a complement-independent inflammatory response in the uteroplacental site by signalling through Toll-like receptor 4 (TLR4) on extravillous trophoblasts [53]. Activation of TLR4 by aPLs results in the secretion of pro-inflammatory chemokines growth-related oncogene α and interleukin-8, as well as the potent pro-inflammatory cytokine interleukin-1β [53]. The conditioned media from extravillous trophoblasts treated with aPLs induces apoptosis in untreated extravillous trophoblasts, suggesting that the pro-inflammatory mediators produced in response to TLR4 activation may act in an autocrine or paracrine fashion to cause extravillous trophoblast death in vitro [53]. Failure of extravillous trophoblasts to invade and remodel the uterine spiral arteries could result in reduced placental blood supply and thus pregnancy morbidity.

In addition to the experiments performed in vitro or with mouse models, workers have also attempted to understand the underlying cause of obstetric morbidity in aPL-positive women by investigating the histopathology of placentae from these women. De Wolf and colleagues were the first to publish a histological report of a placenta from a pregnancy affected by aPLs. Fifty percent of the placenta in that report was infarcted, and several spiral arteries supplying the placenta with maternal blood were occluded by intraluminal thromboses [54]. The observation of spiral artery thrombosis and placental infarction in this and other early case reports [55], [56] was the impetus for the theory that thrombotic occlusion of the spiral arteries, impairment of placental blood supply, placental infarction, and a consequent reduction in materno-foetal exchange, was the cause of foetal demise in pregnancies affected by aPL. Since these initial case reports, many more workers have investigated the histopathology of placentae from aPL-positive women in larger case-control studies. However, not all of the placentae from aPL-positive women in these larger studies exhibited signs of spiral artery thrombosis or placental infarction, suggesting that the early theory implicating these lesions in foetal demise was overly simplistic and that other pathological processes may contribute to pregnancy morbidity in women with aPLs.

The prevalence of various histopathological lesions in the placentae of women with aPL has been difficult to reconcile from the existing literature since different case–control studies have investigated different histopathological lesions in several diverse groups of women with results that are often contradictory. This systematic review was undertaken in order to generate a complete and coherent picture of the placental features associated with aPLs in an attempt to better understand the underlying pathological processes that take place in aPL-affected pregnancies. It was hypothesised that, in line with the current evidence from in vitro work and mouse models, this would uncover a placental aPL fingerprint that involved thrombotic or inflammatory lesions or evidence of trophoblast dysfunction and death.