November 1998 Volume 179 Number 5 Etiology and pathogenesis of preeclampsia: Current conceptsGustaaf A. Dekker, MD, PhD [MEDLINE LOOKUP] Baha M. Sibai, MD [MEDLINE LOOKUP]

Amsterdam, The Netherlands, and Memphis, Tennessee

Sections

Abstract

Uteroplacental circulation

Vascular tone

Pathogenesis of preeclampsia

Etiology of preeclampsia

Preeclampsia as a genetic disease

Underlying disorders associated with preeclampsia

Unifying hypothesis (Fig 1)

References Publishing and Reprint Information Articles with References to this Article

Previous article in Issue Next article in Issue View print version (PDF) Drug links from Mosby's DrugConsult Genetic information from OMIM Citation of this Article View on PubMed Download in citation manager format Download in Medlars format Related articles in PubMed

AbstractTOP

The etiology of preeclampsia is unknown. At present, 4 hypotheses are the subject of extensive investigation, as follows: (1) Placental ischemiaIncreased trophoblast deportation, as a consequence of ischemia, may inflict endothelial cell dysfunction. (2) Very low-density lipoprotein versus toxicity-preventing activityIn compensation for increased energy demand during pregnancy, nonesterified fatty acids are mobilized. In women with low albumin concentrations, transporting extra nonesterified fatty acids from adipose tissues to the liver is likely to reduce albumins antitoxic activity to a point at which very-low density lipoprotein toxicity is expressed. (3) Immune maladaptationInteraction between decidual leukocytes and invading cytotrophoblast cells is essential for normal trophoblast invasion and development. Immune maladaptation may cause shallow invasion of spiral arteries by endovascular cytotrophoblast cells and endothelial cell dysfunction mediated by an increased decidual release of cytokines, proteolytic enzymes, and free radical species. (4) Genetic imprintingDevelopment of preeclampsia-eclampsia may be based on a single recessive gene or a dominant gene with incomplete penetrance. Penetrance may be dependent on fetal genotype. The possibility of genetic imprinting should be considered in future genetic investigations of preeclampsia. (Am J Obstet Gynecol 1998;179:1359-75.)

(Click on a term to search this journal for other articles containing that term.)

Key words: Preeclampsia, endothelium, etiology, immune, gene

Shallow endovascular cytotrophoblast invasion in the spiral arteries and endothelial cell dysfunction are 2 key features in the pathogenesis of preeclampsia, yet the etiology of preeclampsia is still unknown. Currently, 4 hypotheses are favored: placental ischemia, very low-density lipoproteins versus toxicity preventing activity, immune maladaptation, and genetic imprinting. For the sake of clarity, these 4 hypotheses are described separately; however, it should be stressed that these 4 hypotheses are not mutually exclusive but probably interactive.

Before discussing the pros and cons of these 4 hypotheses, we will briefly review the literature on uteroplacental circulation, vascular tone regulation in normal pregnancy, and pathogenesis of the clinical syndrome.

Uteroplacental circulationTOP

During the early weeks of gestation, cytotrophoblast cells stream out of the tips of the anchoring villi and penetrate the trophoblast shell and the overlying syncytiotrophoblast to form cytotrophoblast columns that develop into the cytotrophoblast shell. Trophoblast cells continue to migrate into the decidua and eventually colonize the placental beds myometrium. When the cytotrophoblast shell contacts the spiral arteries openings, trophoblast cells stream into their lumina, where they form intraluminal plugs.1-3 Endovascular trophoblast cells replace the endothelium of spiral arteries and then invade the media, with resulting destruction of the medial elastic, muscular, and neural tissue. Trophoblast cells become incorporated into the vessel wall, and in the final stages the endothelial lining is reconstituted.1-3 Zhou et al4 demonstrated that endovascular cytotrophoblasts normally transform their adhesion receptor phenotype so as to resemble the endothelial cells they replace, and they were subsequently successful in demonstrating that preeclampsia is associated with a failure of cytotrophoblasts to mimic a vascular adhesion phenotype. Vascular changes may actually precede trophoblast invasion, suggesting that interstitial trophoblast plays a role in inducing early disruption (probably by secreting proteolytic enzymes).3 These physiologic changes create a low-resistance arteriolar system and an absence of maternal vasomotor control, which allows the dramatic increase in blood supply to the growing fetus. During early stages of implantation, cytotrophoblast plugs may act as valves regulating blood flow in the intervillous space and protect the embryo from forceful maternal blood flow. Results of color Doppler studies suggest that the presence of continuous intervillous space flow in the first trimester is associated with a complicated pregnancy outcome and that a true intervillous space blood flow is established only by about 12 weeks gestation in normal pregnancies.4-6In preeclampsia and some cases of intrauterine growth restriction (IUGR), physiologic changes in the spiral arteries are confined to the decidual portion of the arteries; about 30% to 50% of the placental beds spiral arteries escape entirely from endovascular trophoblast invasion.1,2,7 Myometrial segments remain anatomically intact and undilated, and adrenergic nerve supply to the spiral arteries remains intact.2,7 Intraluminal cytotrophoblast in spiral arteries is often identified in preeclampsia and IUGR, unlike in normal pregnancies.2,7 Many vessels are occluded by atherosis, with an accumulation of lipid-laden macrophages and a perivascular mononuclear cell infiltrate. Acute atherosis and the associated thrombosis may cause placental infarctions. Lipoprotein-a deposition is often found in association with atherosis.8Vascular endothelial growth factor (VEGF) and the vascular adhesion phenotype of invading cytotrophoblasts.VEGF, a growth factor involved in angiogenesis, with a specific action on endothelial cells, is widely expressed at the fetomaternal interface.9 On implantation, expression is dominant in the uterine epithelial cells. Subsequently, expression increases in the decidual macrophages, which remain the principal source of VEGF. First-trimester human intermediate and extravillous trophoblast has been found to express the flt receptor for VEGF. VEGF is up-regulated by hypoxia10; neovascularization induced by hypoxia corrects tissue oxygenation, and VEGF release will be down-regulated. This homeostatic mechanism may have particular relevance in the first trimester when oxygen tension of trophoblast villi is particularly low.4-6,10-12 VEGF may be involved in inducing the expression of endothelial markers by invading cytotrophoblasts. VEGF has been shown to induce the expression of those integrins on endothelial cells that are associated with angiogenic invasion; these integrins are the same as those described by Zhou et al4 as being part of the de novo expression of endothelial characteristics by invading cytotrophoblasts.

Decidual extracellular matrix and cell surface adhesion molecules.During invasion, trophoblast cells are not cytolytic but secrete enzymes that affect the extracellular matrix. Immunohistochemical studies have localized plasminogen activator, as well as metalloproteinases, which include collagenases within the extravillous trophoblast.3,13,14 Metalloproteinase activity is influenced by a variety of mediators. Secretion of metalloproteinases is plasmin dependent but is inhibited by -human chorionic gonadotropin, which would favor an autocrine regulatory system. However, interleukin-l (IL-1) is stimulatory, and because trophoblast cells release IL-1 and also express the IL-1 receptor, there may be an autocrine stimulatory loop.15 Another potential candidate for stimulation of type IV collagenase is tumor necrosis factor- (TNF-).3Cell migration is dependent on adhesion to extracellular matrix proteins for anchorage and traction. In addition to mediating adhesion, extracellular matrix binding to specific transmembrane receptors also affects cell growth and differentiation.16,17 Also, trophoblast behavior during implantation is influenced by decidual extracellular matrix proteins.18,19 Cells bind to extracellular matrix proteins by appropriate surface receptors called adhesion molecules.20 The 4 major families of adhesion molecules are integrins, cadherins, immunoglobulin superfamily, and selectins. E-Cadherin is involved in differentiation of cytotrophoblast into syncytiotrophoblast,21 whereas integrins mediate adhesion to the extracellular matrix.20 Integrins are transmembrane glycoproteins consisting of noncovalently associated - and -subunits.22 Integrin ligand specificity is determined by combination of the various - and -subunits.22-26 Thus onset of migratory or invasive behavior of trophoblast is associated with a shift in integrin expression, thereby changing the binding tendency of trophoblast from a basal lamina (laminin) to a stromal type (fibronectin) of matrix.3,27-31Integrin switching by invasive cytotrophoblasts is abnormal in preeclampsia. Invading trophoblast within the placental bed of preeclamptic patients fails to down-regulate 4 and to make the switch from 6 to 5 and 1 integrin subunits.25,27-30 Pijnenborg et al32 observed lower trophoblast attachment on the extracellular matrix proteins fibronectin and vitronectin in preeclamptic pregnancies. This probably reflects differences in trophoblast integrin expression. In contrast, Divers et al33 failed to find any difference in trophoblast integrin expression between normal and preeclamptic pregnancies. The abnormal cytotrophoblast differentiation in preeclampsia is not necessarily a primary feature. An alternative explanation is an abnormal interaction between trophoblast and decidual lymphoid cells.31 In a monkey model hypoxia was demonstrated to be associated with increased invasion.34 However, it is uncertain whether this phenomenon is mediated by a direct effect of hypoxia on cytotrophoblast integrin expression or reflects increased VEGF action.4 Vascular toneTOP

In normal pregnancy there is an 8-fold to 10-fold increase in prostacyclin. Thromboxane A2 (TxA2) biosynthesis is also increased throughout normal pregnancy; the net result, however, is a biologic dominance of prostacyclin over TxA2.35,36 The dominance of prostacyclin provides an explanation for vascular refractoriness to angiotensin II in normal pregnancy, although its role as a circulating hormone is still controversial because of its evanescent nature.35-37 However, prostaglandins may not be the major vasodilators of pregnancy.38,39 A number of peptide regulatory factors (cytokines, binding proteins, growth factors) released in an appropriate cortico-steroid environment play an integral part in this mediation.40-42 Nitric oxide is probably an important messenger, but certainly not the only one, in inducing vasodilation in normal pregnancy.42-47 In pregnant animals chronic inhibition of nitric oxide synthesis reverses refractoriness to angiotensin and vasopressin and eventually produces a preeclampsia-like syndrome.48 Nitric oxide is probably not the only explanation for the marked vasodilatation of normal pregnancy. Pascoal and Umans49 recently provided strong evidence for the presence of an endothelium-derived hyperpolarizing vasodilatory factor in causing pregnancy-associated vasorelaxation.

Nitric oxide is more important than prostacyclin in maintaining low vascular tone in fetoplacental vessels.50-52 Immunohistochemical studies to localize endothelial nitric oxide synthetase in term placentae found strong staining patterns in syncytiotrophoblast and absence of staining in endothelial cells of small fetoplacental vessels and the cytotrophoblast.53 Negligible nitric oxide production in the placental bed is unlikely to contribute substantially to uteroplacental vasodilation.54 Additionally, recent studies found significantly higher inducible nitric oxide synthetase activity in basal plate and villous tissue compared with the activity of endothelial nitric oxide synthetase.54,55 Pathogenesis of preeclampsiaTOP

In preeclampsia, absence of normal stimulation of the renin-angiotensin system, despite hypovolemia, and increased vascular sensitivity to angiotensin II and norepinephrine can be explained by a biologic dominance of TxA2 over prostacyclin.35-37,56,57 A reduction in urinary excretion of prostacyclin metabolites precedes the development of clinical disease, whereas TxA2 biosynthesis is increased in preeclampsia.58 Increased TxA2 production in preeclampsia is largely derived from platelets35 and the placenta.59 Placental prostacyclin production is reduced.60 Although the increased TxA2-to-prostacyclin ratio allows an explanation for vasoconstriction, platelet destruction, and reduced uteroplacental blood flow, evidence is mounting that preeclampsia is not simply a state of prostacyclin deficiency.57,61 Increased levels of factor VIIIrelated antigen, total and cellular fibronectin, thrombomodulin, endothelin, growth factor activity, a disturbance of the tissue plasminogen activator/plasminogen activator inhibitor balance, and the prostacyclin/TxA2 balance all support the hypothesis that a more global endothelial cell dysfunction is intimately involved in the pathogenesis of preeclampsia.62-73 Morphologic evidence of endothelial cell injury is provided by glomerular endotheliosis and ultrastructural changes in the placental bed, uterine boundary vessels, and other parts of the circulation.74-76 In in vitro experiments preeclamptic sera do not cause actual endothelial cell damage but rather activate specific endothelial cell metabolic pathways,77 which may be related to the disturbances in vascular function encountered in vivo.78 Results of studies on actual nitric oxide production in preeclampsia have yielded controversial results: Plasma nitrite and nitrate levels have been found to be decreased, unchanged, and increased as compared with the levens in normotensive control subjects.79-84 However, the best indicator (in vivo) of overall nitric oxide production is the steady-state urinary excretion of nitrites, and Davidge et al85 found urinary excretion, but not plasma levels, of these nitric oxide metabolites to be decreased in preeclampsia. Plasma concentrations of endogenous nitric oxide synthetase inhibitors are raised in patients with preeclampsia as compared with the levels in patients with normal pregnancies or patients with gestational hypertension.86,87 Nitric oxide donors cause a significant decrease in uterine artery resistance in patients with an elevated uterine artery resistance index.88In preeclampsia, endothelial cell dysfunction and platelet aggregation precede the increase in thrombin and fibrin formation.89,90 An inadequate production of either antiaggregatory prostacyclin or nitric oxide, or both, provides an attractive hypothetical explanation for surface-mediated platelet activation occurring on the inner lining of spiral arteries. Platelets adhere to and release and dense granule constituents, specifically, TxA2 and serotonin, contributing to platelet aggregation and inducing fibrin formation. Zeeman and Dekker61 suggested that if there are still intact endothelial cells left in the spiral arteries the increased platelet-derived serotonin levels91 may interact with endothelial S1 receptors, which may induce a partial recovery of endothelial prostacyclin and nitric oxide release.92 Local prostacyclin may stimulate the uteroplacental renin-angiotensin system and thus induce the release of angiotensin II, which may improve uteroplacental perfusion by increasing maternal blood pressure (perfusion pressure) and by forming an extra stimulus for prostacyclin and nitric oxide release by the uteroplacental vessels.61,93In preeclampsia and IUGR the fetoplacental unit is also characterized by a prostacyclin/TxA2 imbalance.94 Concerning fetoplacental nitric oxide synthesis, Morris et al54 found syncytiotrophoblast nitric oxide release, as judged by nitric oxide synthetase activity, to be decreased in preeclampsia as compared with normal placentas. Wang et al95 found no differences in nitric oxide production, as judged by nitrite levels, between normal and preeclamptic placentas. Ghabour et al96 reported that syncytiotrophoblast endothelial nitric oxide synthetase immunostains appeared primarily apical in location and diffuse in distribution in preeclamptic placentas but primarily basal and punctate in normal placentas. In the villous trophoblast differentiation of cytotrophoblast into syncytiotrophoblast is associated with endothelial nitric oxide synthetase expression.97 Thus alterations in endothelial nitric oxide synthetase expression in preeclamptic placentas may be a result of placental damage and repair processes rather than the primary mediator.

Etiology of preeclampsiaTOP

Endothelial cell activation or dysfunction appears to be the central theme in the pathogenesis of preeclampsia, but what causes these endothelial changes in preeclampsia? Four hypotheses are currently the subject of extensive investigation. For the sake of clarity these hypotheses are described separately; however, it should be stressed that they are not mutually exclusive but probably interactive.

Placental ischemia.According to the Oxford group of researchers, preeclampsia is a 2-stage placental disease. The first stage is the process that affects the spiral arteries and results in a deficient blood supply to the placenta. The second stage encompasses the effects of the ensuing placental ischemia on both the fetus and the mother.98-100 Syncytiotrophoblast microvillous membranes from normal and preeclamptic placentas were found to suppress endothelial cell proliferation to a similar extent and disrupted endothelial monolayers to form a honeycomb-like pattern.98 Deportation of placental tissue, a feature also present in normal pregnancy, is increased in established preeclampsia. Although deportation of intact trophoblasts could, at least in theory, exert a systemic effect by impairing the substantial pulmonary nitric oxide production, it is (according to the Oxford group) the increased microdeportation of syncytiotrophoblast microvillous membrane particles that is probably more related to the systemic endothelial cell dysfunction in preeclampsia.99,100 The increased syncytiotrophoblast deportation in preeclampsia is probably explained by the presence of syncytial sprouts that may be elongated on long pedicles.99 Cytotrophoblast proliferation and formation of these outgrowth lesions of syncytiotrophoblast may represent evidence of placental repair mechanisms. Preeclampsia is associated with glycogen accumulation in syncytiotrophoblast,101 which could be another expression of increased syncytiotrophoblast deportation and increased cytotrophoblast proliferation. Placental villous cytotrophoblasts produce TxA2, the amount decreasing as they mature into syncytiotrophoblasts.102 Tsukimori et al103 found a syncytiotrophoblast factor that was cytotoxic to endothelial cells to be present in preeclamptic placentas but not in normal ones.

Placental ischemia as the cause of endothelial cell dysfunction in preeclampsia is an attractive concept, but several concerns are conceivable that may dispute its validity. First, if placental ischemia is the actual cause of endothelial cell dysfunction, one would expect a closer correlation between the presence of maternal and fetal components of the disease. However, this is not the usual finding in clinical practice. Next, the chronology of the maternal components of the syndrome appear not to fit with the placental ischemia hypothesis. The findings of some recent studies suggest that endothelial cell involvement is already present in the first trimester,104,105 which is difficult to reconcile with the first conceivable stages of placental ischemia. The normal environment for trophoblast in the first trimester is low in oxygen,10-12 and this relative hypoxia is likely to be of physiologic importance because signs of increased intervillous flow are associated with adverse pregnancy outcome.6 Low oxygen also affects trophoblast function later in pregnancy, and the fact that hypoxia-induced syncytiotrophoblast and cytotrophoblast changes in third-trimester placentas resemble the histologic changes in preeclampsia suggests that the placenta may indeed be relatively hypoxic in the end stage of a preeclamptic pregnancy.106 The basic problem with the Oxford hypothesis is the fact that the studies are confined to patients with established disease. Thus increased syncytiotrophoblast deportation, which is of pivotal importance in this hypothesis, is likely to be just a feature of end-stage disease. Nevertheless, this phenomenon might be clinically relevant because it could be involved in causing the extreme refractory hypertension and vasospasm, which are so common in the terminal stages of severe preeclampsia-eclampsia.

Very low-density lipoproteins (VLDL) versus toxicity-preventing activity.In preeclampsia, circulating free fatty acids (FFA) are already increased 15 to 20 weeks before the onset of clinical disease.107 Sera from preeclamptic women have both a higher ratio of FFA to albumin and increased lipolytic activity resulting in enhanced endothelial cell uptake of FFA, which are further esterified into triglycerides.108,109 Among the FFA, oleic, linoleic, and palmitic acids have been found to be increased by 37%, 25%, and 25%, respectively. Linoleic acid reduces thrombin-stimulated prostacyclin release by 30% to 60% and oleic acid by 10% to 30%, whereas palmitic acid has no effect. Incubation with linoleic acid reduces endothelial cyclic guanosine monophosphate levels by 70% and the ability to inhibit platelet aggregation by 10% to 45%.110 Plasma albumin exists as several isoelectric species, which range from isoelectric point (pI) 4.8 to pI 5.6. The more FFA are bound to albumin, the lower the pI. Plasma albumin exerts a toxicity-preventing activity if it is in the pI 5.6 form. Because higher ratios of FFA to albumin cause a shift from the pI 5.6 to the pI 4.8 form of albumin, patients with preeclampsia will have lower amounts of protective toxicity-preventing acivity (pI 5.6) than normotensive pregnant women. A low ratio of toxicity-preventing activity to VLDL would result in cytotoxicity and triglyceride accumulation in endothelial cells.111 According to Arbogast et al,111 pregnancy increases energy demands, which are reflected by increasing VLDL concentrations throughout pregnancy. In women with low albumin concentrations, the burden of transporting extra FFA from adipose tissue to the liver is likely to reduce the concentration of toxicity-preventing activity to a point at which VLDL toxicity is expressed, leading to endothelial cell injury.

Similar to the concerns with regard to the placental ischemia hypothesis, it is difficult to reconcile an exaggerated energy demand occurring so early that it will cause a severe VLDL/toxicity-preventing activity imbalance and subsequently endothelial cell dysfunction in the first trimester. Thus FFA and endothelial cell triglyceride accumulation are early and probably very relevant features in preeclampsia, but they are not necessarily caused by an exaggerated energy demand.111,112 Increased FFA levels and endothelial cell triglyceride accumulation in preeclampsia may reflect cytokine-mediated oxidative stress, caused by either ischemia-reperfusion mechanisms or activated leukocytes, or both.113Immune maladaptation.In humans, organ transplants will be rejected if there are differences between donor and recipient with respect to major histocompatibility complex (MHC) genes (ie, human leukocyte antigens [HLA]). It has been found that trophoblast cells in contact with maternal blood are negative for MHC class I and II alloantigens, whereas those in contact with maternal tissue are often MHC class I positive. Hence first-trimester syncytiotrophoblast and noninvasive villous cytotrophoblast do not express MHC class I alloantigens, but rather the extravillous cytotrophoblast at the tips of the cell columns and in spiral arteries is MHC class I positive.ll4-ll7 The HLA class I gene family encodes cell-surface glycoproteins that include the highly polymorphic transplantation molecules HLA-A, HLA-B, and HLA-C, which show widespread tissue expression and function in the presentation of autologous peptide antigens to T cells. At least 3 additional class I genes (HLA-E, HLA-F, and HLA-G, known as nonclassical [class Ib] genes) have been identified that are highly homologous to classical HLA genes. HLA-E and HLA-F are expressed in many fetal and adult tissues. By contrast, HLA-G is only expressed by the extravillous trophoblast at the maternofetal interface, where classical class I and II antigens are absent; this restricted expression suggests that HLA-G plays a role in the immune tolerance of the semiallogenic fetus by the mother.114-117 An important feature of HLA-G is that its gene locus appears to exhibit a very limited degree of polymorphism, which is not likely to affect unduly the sequence of expressed proteins. However, van der Ven and Ober118 presented evidence of many distinctive alleles in the African-American population and the presence of subtle HLA-G mutations in African-American neonates with IUGR, which appears to contrast with the finding that HLA-G shows very little polymorphism in Caucasian persons.

The decidua contains an abundance of cells of bone marrow origin as is demonstrable by their cell surface antigen expression (CD45). These antigens are designated by their CD number, an arbitrary number assigned to each Cluster of Differentiation. In the late secretory phase of the endometrium an unusual population expressing CD56, a marker of peripheral blood large granular lymphocytes, becomes dominant. These uterine large granular leukocytes resemble natural killer (NK) cells but do not express such strong NK cell activity as peripheral blood NK cells. About 75% of decidual cells express CD45, reflecting their bone marrow origin (ie, 45% uterine large granular leukocytes, 19% macrophages, and 8% T cells). As pregnancy progresses, the number of macrophages and T cells remains constant but the number of uterine large granular leukocytes declines dramatically.119-121Cells of the NK lineage, such as uterine large granular leukocytes, kill target cells that are deficient in MHC expression rather than recognizing the presence of foreign MHC. This is the basis of the missing self hypothesis for NK cytolysis.122-124 Therefore simple lack of classical class I molecules on trophoectoderm and later on invading cytotrophoblast would seem to be inadequate to explain their protection from NK(-like) cells. The presence of a nonpolymorphic nonclassical class I antigen, HLA-G, on the surface of invasive cytotrophoblast cells may be an explanation for this conundrum.125 HLA-G is indeed capable of inhibiting NK activity of uterine large granular leukocytes against trophoblast cells at the maternofetal interface.114,120,123 Thus the remarkable absence of regular HLA expression and the presence of HLA-G on invasive cytotrophoblast are likely to be of pivotal significance in protecting trophoblast from effective maternal immune recognition or cytotoxic attack.115-117 Throughout gestation, extravillous cytotrophoblast cells retain the ability to up-regulate HLA-G expression. Apparently they must avoid maternal immune surveillance constantly. In contrast, cytotrophoblast cells only express invasive characteristics transiently (ie, only in early gestation are they characterized by enhanced metalloproteinase expression and integrin switching).126 Colbern et al127 observed a lower level of messenger ribonucleic acid for HLA-G in chorionic tissue in preeclampsia than in normal pregnancy. However, normalization to messenger ribonucleic acid for cytokeratin revealed that this was caused by a smaller number of trophoblast cells present in preeclamptic placentas, thus showing that these findings were only a reflection of the well-established reduced trophoblast invasion in preeclampsia. Recent studies suggest that extravillous trophoblast cells also express HLA-C. It may be that the coordinated expression of both HLA-G and HLA-C is required to provide a set of trophoblast HLA class I molecules, which act together for optimal NK recognition and to prevent lysis by these cells.114,116An alternative biologic role for uterine large granular lymphocytes and other decidual CD56 cells has recently been consideredthat trophoblast growth and invasion may be dependent on appropriate cytokines being produced by these cells in response to HLA-G expressed on cytotrophoblast cells.128 Hence decidual leukocyte activity could actually be beneficial to trophoblast growth and function by a phenomenon named immunotrophism.129 In addition, colony-stimulating factors are produced by decidual macrophages and by the developing placenta itself. The amount increases with implantation.130 Colony-stimulating factors stimulate the endometrial macrophage population, trophoblast human placental lactogen, and human chorionic gonadotropin synthesis and thus appear to be intimately involved in decidual trophoblast interactions during early pregnancy.131 Several cytokines are now known to be integral to early pregnancy development. Granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-1, TNF-, interferon-(IFN-), and colony-stimulating factor-1 (CSF-1) have all been implicated in blastocyst attachment and in trophoblast implantation, proliferation, and invasion. Because TNF-, IFN-, IFN-, IFN-, and transforming growth factor (TGF)-l are produced in the uteroplacental unit and inhibit trophoblast deoxyribonucleic acid synthesis, they may take part in regulating trophoblast growth. Conversely, IL-1, GM-CSF, and IL-6 are thought to have trophoblast growth-stimulating effects. It should be emphasized that the majority of these cytokine effects occur in a close and complex interaction with cortico-steroids and prostaglandins.l19, 132-134In 1986, 2 distinct and mutually inhibitory types of T-helper cells were described.135 The first type of cell, termed Th1, secretes IL-2, IFN-, and lymphotoxin. This contrast with Th2 cells, which secrete IL-4, IL-4, IL-6, and IL-10.136 Th1 cytokines are associated with cell-mediated immunity and delayed hypersensitivity reactions, whereas Th2 cytokines foster antibody responses and allergic reactions. Because Th1 cytokines are considered to be harmful to pregnancy and Th2 cytokines (IL-10) can down-regulate production of Th1 cytokines, it has been proposed that successful pregnancy is a Th2 phenomenon.137 This hypothesis is consistent with the observation that cell-mediated autoimmune conditions such as rheumatoid arthritis are often improved during pregnancy, whereas disorders associated with autoantibodies may be worsened by pregnancy.138 Several substances such as prostaglandin E2, TGF-, GM-CSF, and IL-10139-141 play a part in the immunoendocrine network in pregnancy maintenance. PGE2 has many immunosuppressive properties, including the inhibition of all cells of the immune system.140 IL-10 potentially has 2 mechanisms by which it may inhibit immune function: (1) directly as a cytokine synthesis inhibitory factor and (2) indirectly as an inducer of adrenocorticotropic hormone. Although these data are intriguing, they must be considered in light of the successful reproduction of IL-10 knock-out mice.142In addition to cytokines, fetal and placental growth is at least partially regulated by insulin-like growth factors (IGF) and insulin-like growth factorbinding proteins (IGFBP-1 or placental protein 12). Trophoblast substances such as human placental lactogen stimulate an increasing production of IGF-I by the maternal liver, whereas, on the other hand, IGF-I binding to trophoblast cells changes during trophoblast differentiation.143 Fetal and maternal IGF-I levels are positively correlated with fetal weight. IGFBP-1, which is produced by decidual stromal cells, impairs trophoblast growth by inhibiting IGF-I. Another explanation may be that IGFBP-1 also contains an integrin arginineglycineaspartic acid recognition sequence that can bind to the 4D1 integrin (fibronectin receptor).144 The presence of elevated maternal levels of IGFBP-1, at 16-24 weeks gestation, in future preeclampsia and IUGR suggests that decidual overproduction of IGFBP-1 may be at least partially involved in the causation of abnormal placentation.143-145 In contrast, de Groot et al146 found midtrimester plasma IGFBP-1 concentrations to be significantly lower in women who later had preeclampsia compared with control subjects with normal pregnancies. At the moment it is not clear whether these contradictory findings can be explained by assay differences.

Epidemiologic evidence suggests that immune mechanisms ought to be involved in the etiology of preeclampsia.147-149 Genuine preeclampsia is primarily a disease of first pregnancies. A previous normal pregnancy is associated with a markedly lowered incidence of preeclampsia. Even a previous abortion provides some protection. Multiparitys protective effect, however, is lost with change of partner; thus preeclampsia may be a problem of primipaternity rather than primigravidity.148 In a recent article Trupin et al149 studied 5068 nulliparous and 5800 multiparous women, 573 of whom had new partners. The preeclampsia incidence in nulliparous women (3.2%) and multiparous women with changed paternity (3%) was found to be similar, as contrasted with the significantly lower incidence of preeclampsia (1.9%) in multiparous women with no change in partners. Analogous to altered paternity, artificial donor insemination and oocyte donation result in an increased incidence of preeclampsia.150 Additionally, the length of time of unprotected sexual cohabitation before conception appears to be inversely related to the incidence of pregnancy-induced hypertensive disorders.151 There are numerous reports of immunologic phenomena in preeclampsia. Beginning early in the second trimester, women in whom preeclampsia develops later have a significantly lower proportion of T-helper cells compared with the proportion in those women who remain normotensive.152 Antibodies against endothelial cells have been found in 50% of women with preeclampsia versus 15% of control women.153 Deposition of immunoglobulins (IgM), complement (C3), and fibrin has been noted in the walls of preeclamptic spiral arteries, especially in acute atherosis, in which substantial granular deposits of IgM can often be seen. Immunologic pathogenesis of acute atherosis is suggested by morphologic similarities to lesions seen in allograft rejection.154 However, acute atherosis usually occurs in those arteries that have not undergone physiologic changes mediated by trophoblast invasion; thus, if a transplantation analogy is pursued, it is necessary to consider that the fetus is rejecting the mother.154What is the link between immune maladaptation and endothelial cell activation in preeclampsia? The decidua contains an abundance of cells of bone marrow origin. These cells, when activated, can release a host of mediators that may interact with endothelial cells.

Elastase.Activated neutrophils release elastase and other toxic proteases, which can destroy the integrity of endothelium. Plasma elastase levels are significantly higher in preeclampsia than in normal pregnancy and correlate with elevated levels of endothelin and factor VIIIrelated antigen, supporting their involvement in causing endothelial cell dysfunction in preeclampsia. The number of elastase-positive decidual neutrophils is significantly greater in preeclampsia than in normal pregnancies and correlates with plasma uric acid.155,156Cytokines.So far, most cytokine research has been concerned with TNF and IL-1 because their endothelial effects resemble changes described in preeclampsia. TNF and IL-1 enhance the generation of thrombin, platelet-activating factor, factor VIIIrelated antigen, and plasminogen activator inhibitor-l, endothelial cell permeability, and expression of intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) and increase inducible nitric oxide synthetase activity and levels of a variety of prostaglandins, at the same time that endothelial nitric oxide synthetase activity is down-regulated.157,158 So far, most studies reported found an increase in TNF- levels in preeclampsia, although it is still controversial whether this increase in TNF- occurs after the syndrome is detected clinically and whether this increase only occurs in preeclamptic pregnancies complicated by IUGR.159-161 TNF- detection is problematic as a result of a short half-life and possible interference of its soluble receptors. It may be deceptive to draw conclusions from plasma TNF levels. Soluble TNF receptors are more reliable markers for TNF activity. There are 2 specific receptors for TNF- of 75 and 55 kd molecular weight (p75 and p55, respectively). Each can be shed as soluble protein that specifically binds TNF. Levels of soluble TNF receptors are known to increase after an increase in TNF-, bind TNF-, and thus provide a protective mechanism against excessive TNF- production. Because TNF receptors have a longer half-life than TNF-, they are footprints of TNF- activity. The p55 and p75 soluble TNF receptor levels are significantly elevated in preeclampsia as compared with uncomplicated pregnancy.162,163 Thus the rise in TNF- levels reported in preeclampsia is probably real.164 The source of this excessive production may be activated (decidual) leukocytes or the placenta itself, because syncytiotrophoblast contains TNF- messenger ribonucleic acid.165 Preeclampsia shows some aspects of an acute-phase response, which may be caused by the elevated IL-6 levels.166 Changes in plasma proteins, which include increases in plasma ceruloplasmin, l-antitrypsin, and haptoglobin, hypoalbuminemia, and reduced plasma transferrin are characteristic of an acute-phase reaction, as are certain alterations in complement activity in preeclamptic sera. Also, Il-6 has definite effects on endothelial cells such as increased permeability, stimulation of platelet-derived growth factor synthesis, and impairment of prostacyclin synthesis. Oxygen free radicals are known to induce endothelial IL-6 synthesis. Il-6 production is thought to be integrally related to TNF-. Whereas IL-6 has direct negative feedback on TNF- production, its production in decidua and trophoblast is increased by TNF-a and IL-1.167 IFN- levels are increased in preeclampsia.168 IFN- up-regulates endothelial cell ICAM-1 expression and IL-6 synthesis, which may be another explanation for the increased plasma IL-6 levels in preeclampsia. IL-8 and GM-CSF levels have been found to be similar in normal and preeclamptic pregnancies.166Dysfunctional endothelial cells undergo activation with production of cell surface proteins that mediate adherence of inflammatory cells. This inductive process is mediated by cytokines produced by inflammatory cells and activated endothelial cells. A number of so-called leukocyte-endothelial adhesion molecules have been identified, including E-selectin, VCAM-1, and ICAM-1. ICAM-1 and VCAM-1 have a wide tissue distribution, whereas E-selectin is only expressed by endothelial cells. Preeclampsia is associated with increased maternal serum levels of soluble VCAM-1, E-selectin, and probably also ICAM-1.168,169 This increase may be an early event.168 The increased IL-6 and IL-1 receptor antagonist (reflecting increased IL-1 concentrations) levels in preeclampsia correlate with elevated VCAM-1 concentrations.166 The increase in these soluble cell adhesion molecules reflects increased expression of these molecules on endothelial cells and thereby explains leukocyte activation in preeclampsia.156 Abnormal endothelial cellleukocyte interaction in preeclampsia is confined to maternal circulation, as is demonstrated by similar expression of cell adhesion molecules in placental vessels and elastase levels in fetal plasma in preeclampsia as compared with normal pregnancy.156,170Oxygen free radicals.Preeclampsia is associated with oxidative stress, defined as an imbalance between prooxidant and antioxidant forces in favor of prooxidants, leading to potential cell or tissue damage.171 Oxygen free radicals can lead to the formation of lipid peroxides. Once lipid peroxidation is initiated, it becomes self-propagating. According to some investigators, oxidized low-density lipoproteins may have a half-life of 3 hours in the circulatory system, suggesting the still controversial possibility that they may function as circulating compounds.171-173 Because lipid peroxides, measured as stable lipid peroxide metabolites,172 are increased above nonpregnant levels in normal pregnancy, it seems that even normal pregnancy induces a certain degree of oxidative stress. Lipid peroxide levels are further increased in preeclampsia and correlate with an increase in blood pressure but not with perinatal outcome.174-178 Preeclamptic lesions of decidual vessels resemble atherosclerotic lesions, both showing fibrinoid necrosis of the vessel wall and accumulation of lipid-laden foam cells, a hallmark of oxidized low-density lipoproteins.173,174,179 Some,180 but not all,181 investigators found titers of IgG autoantibody to oxidized low-density lipoproteins to be increased in preeclampsia.

Activated leukocytes may provide one of the major sources of oxygen free radicals in preeclampsia. Additionally, activated leukocytes may, by producing certain cytokines, irreversibly convert endothelial cell xanthine dehydrogenase to xanthine oxidase, thus causing endothelial release of oxygen free radicals.182-184 Oxygen free radicals and elastase may work synergistically in causing endothelial cell damage.182,185,186 The placenta is another source of free radical species in preeclampsia; placental lipid peroxide production correlates with the increased placental production of TxA2.175 Metabolism of arachidonic acid generates oxygen free radicals by both lipoxygenase and cyclooxygenase pathways, but the cyclooxygenase pathway generates >1000 times more superoxide than the lipoxygenase pathway.187 Walsh175 hypothesized that neutrophils could be activated as they circulate through the intervillous space by lipid peroxide generated by trophoblast cells. Increased lipid peroxide has also been observed in platelets and red blood cells.188Free radical species cause a type of endothelial cell dysfunction that is similar to changes seen in preeclampsia.174,175 Lipid peroxides stimulate prostaglandin H synthase but inhibit prostacyclin synthase; therefore increased lipid peroxide levels in preeclampsia favor production of platelet-derived TxA2 above vascular prostacyclin synthesis. Free radical species impair endothelial cell nitric oxide production and inhibit macrophage inducible nitric oxide synthetase.189,190 Lipid peroxides alter capillary permeability to proteins186 and may thus be involved in causing edema and proteinuria.175 Lipid peroxides trigger thrombosis by increasing thrombin generation and endothelial plasminogen activator inhibitor-1 release while at the same time decreasing antithrombin and endothelial tissue plasminogen activator release.191 Lipid peroxides alter cell membrane fluidity by increasing incorporation of cholesterol, oxidized fatty acids, and low-density lipoproteins. In preeclampsia this phenomenon has been described for platelets, erythrocytes, and trophoblast cells.188,192 This may partially explain why incubation of endothelial cells with sera from preeclamptic women increases the endothelial cell content of triglycerides, which results in a disturbance of endothelial cell physiologic characteristics.77,108,113Antioxidants are a diverse family of compounds designed to prevent overproduction of and damage caused by free radical species. The picture in preeclampsia is complex, but it is apparent that the disease cannot be characterized as a state of global antioxidant deficiency.193 One apparent paradox of the oxidative stress theory of preeclampsia concerns the antioxidant function of uric acid.183,194 In preeclampsia the ability of plasma to scavenge aqueous radicals is markedly augmented, and most of this increase is caused by the increase in uric acid.193,194 Thus, although increased xanthine oxidase activity may contribute to oxidative stress in preeclampsia, hyperuricemia itself may serve a protective role as antioxidant.194Vitamin E is the major lipid-soluble antioxidant. According to Wang et al,36 maternal concentration ratios of antioxidant vitamin E to lipid peroxides and prostacyclin to TxA2 favor the antioxidant activity of vitamin E and the vasodilator effects of prostacyclin with advancing gestation in normotensive pregnancy, whereas in severe preeclampsia the lipid peroxide levels are increased and the vitamin E levels are decreased; this imbalance correlates with the imbalance of increased TxA2 and decreased prostacyclin177 and with the increase in blood pressure.195 Several authors did not find decreased vitamin E levels in preeclampsia.178,193,196Generation of highly toxic radicals, such as OH, that initiate lipid peroxidation and reactions causing propagation of lipid peroxidation require the presence of a low-molecular-weight pool of iron or other transition metals. Thus a critical group of antioxidants is those designed to bind metal ions in forms that will not generate free radical species, including transferrin, albumin, and ceruloplasmin.197,198 When brain tissue is homogenized in a buffer solution, endogenous lipids undergo rapid peroxidation as a result of transition metals released from cells. The ability of small amounts of serum to inhibit this peroxidation provides an assay for serum antioxidant activity. Peroxidation is inhibited by several iron-chelating agents; thus the assay has a predilection for antioxidants that function by sequestering metals.197-199 Using this assay, Davidge et al199 found an increased serum antioxidant capacity during uncomplicated pregnancy, whereas the antioxidant capacity in preeclampsia was found to be reduced by 50%.

Erythrocyte glutathione levels and glutathione peroxidase activity, another lipid peroxide scavenger system, are increased in preeclampsia in comparison with normal pregnancy; this increase may represent a compensatory response.176 Tissue glutathione peroxidase levels may be different from erythrocyte levels. Indeed, Walsh and Wang200 found decreased glutathione peroxidase levels in placentas from preeclamptic pregnancies as compared with normotensive pregnancies.

Ascorbic acid, -tocopherol, and -carotene are potent antioxidant nutrients. Ascorbic acid has been described as the major front-line water-soluble antioxidant. Ascorbic acid levels are significantly decreased in patients with mild and severe preeclampsia, whereas -tocopherol and -carotene levels are significantly decreased only in severe preeclampsia.201 Thus, in preeclampsia, antioxidant nutrients may be used to a greater extent to counteract free radical-mediated cell disturbances, resulting in a reduction in antioxidant plasma levels. Water-soluble antioxidant nutrients may initially be consumed, followed by lipid-soluble antioxidants. Diminished serum metal-binding antioxidant capacity and placental antioxidants may be of special importance in allowing initiation and propagation of lipid peroxidation in preeclampsia.175,194,199The hypothesis that cytokines, especially TNF-, may contribute to oxidative stress in preeclampsia was reviewed by Stark.202 Mitochondrial injury is the first detectable effect of TNF- in cultured cell lines, whereas some studies found preeclampsia to be associated with a very similar type of mitochondrial damage.74-76 Oxidative disequilibrium in preeclampsia is probably closely linked with cytokine activity, particularly TNF-. Antioxidants selectively inhibit TNF- release because they control the oxidation-reduction status of glutathione peroxidase, which is of major importance as an endogenous modulator of TNF- production. In mitochondria, TNF- subverts part of the electron flow to release oxygen free radicals with subsequent lipid peroxide formation. If these are not removed with subsequent efficient reduction of the oxidized glutathione peroxidase, serious toxic effects ensue.

Immune maladaptation is an attractive hypothesis to explain several of the well-known epidemiologic features of preeclampsia. It should be emphasized that although there is no doubt that patients with preeclampsia manifest immunopathologic characteristics, the question remains as to whether such immunologic abnormality causes or results from preeclampsia. Immunoglobulin, immune complex, complement, and fibrin deposits in decidual vessels may be the effect of occluded vessels rather than the cause; analogous results described in chorionic villous tissue may have resulted from placental ischemia, and the IgM, IgG, and complement deposits in glomeruli in preeclampsia also are not direct proof for an immunologic pathogenesis of these lesions. The immune maladaptation hypothesis is, of course, not mutually exclusive with the placental ischemia hypothesis. Immune maladaptation may be the cause of abnormal implantation and thus evoke placental ischemia. Moreover, placental ischemia provides an alternative explanation for the cytokine changes203 and oxidative stress seen in preeclampsia. Placental cells express erythropoietin, which is the prototype molecule for transcriptional regulation by hypoxia in mammals. TNF- and IL-1 have deoxyribonucleic acid sequences that are homologous or nearly homologous to the hypoxia-responsive enhancer element of the erythropoietin gene, thus providing a potential but as of yet untested molecular link between placental hypoxia and stimulation of cytokine production.203 Preeclampsia as a genetic diseaseTOP

Severe preeclampsia and eclampsia have a familial tendency. Chesley et al204 found a 26% incidence of preeclampsia in daughters of women with preeclampsia but only an 8% incidence in the daughters-in-law. Published reports are consistent with a relatively common allele acting as a major gene conferring susceptibility to preeclampsia. The development of preeclampsia may be based on a single recessive gene or a dominant gene with incomplete penetrance. Multifactorial inheritance is another possibility.205,206 The principal reason for suspecting the existence of a maternal gene as the sole explanation for preeclampsia is data showing a lack of concordance for susceptibility in a small sample (n = 10) of monozygous twins.207 The increased prevalence of preeclampsia in daughters born to an eclamptic mother, as opposed to a noneclamptic pregnancy of the same mother, could indicate an influence of the fetal genotype on susceptibility to preeclampsia.205,206 An example of the impact of the fetal genotype on preeclampsia is the association between HELLP (hemolysis, elevated liver enzymes, and low platelets) syndrome and a rare fetal metabolic disorder208 and the increased incidence of preeclampsia in cases of fetal chromosomal abnormalities (for example, triploid and trisomy 13).209Arngrimsson et al210 provided epidemiologic evidence that the genetic predisposition to preeclampsia may be related to implantation and placentation.

Conflicting data have been reported regarding the involvement of HLA antigens. Several studies revealed an increased HLA compatibility between patients and their mates.205 Kilpatrick et al211 found a significant association between HLA-DR4 and preeclampsia. However, association of preeclampsia with HLA-DR4 has now been excluded by detailed genetic analysis of restriction fragment polymorphisms of HLA-DR4 in informative pedigrees; the association probably indicates an underlying tendency to autoimmune disease with which HLA-DR4 is linked.212,213 In contrast, Peterson et al214 observed that mothers with HLA haplotypes A23/29, B44, and DR7 are at a higher risk for development of preeclampsia and IUGR than mothers without such haplotypes. Humphrey et al215 studied HLA-G deletion polymorphisms in preeclamptic pedigrees and controls and concluded that there is no detectable relationship between susceptibility to preeclampsia or being born of a preeclamptic pregnancy and the HLA genotype. However, as mentioned earlier, Van der Ven and 0ber118 demonstrated the presence of subtle HLA-G mutations in African-American neonates with IUGR.

An amino substitution in the angiotensinogen gene on chromosome 1 (the presence of threonine rather than methionine at residue 235 [T235] in the protein) has been suggested to correlate with the risk of preeclampsia but not with the risk of HELLP syndrome.216 Other studies found no association between the T235 allele and preeclampsia217,218 and suggested that the T235 allele has more to do with a tendency for high blood pressure to develop rather than with preeclampsia.219 Genetic alterations in angiotensin II receptors are currently studied because changes in their expression may be involved in regulating vasodilation in uteroplacental and fetoplacental vessels.220,221 Involvement of (genetic) TNF-mediated mitochondrial dysfunction in the etiology of preeclampsia158,203 is supported by the increased incidence of severe preeclampsia in a family with a failure of the reduced nicotinamide adenine dinucleotide ubiquinone oxidoreductase step in the mitochondrial electron chain.222 Furui et al223 observed reduced expression of mitochondrial genes in placental dysfunction in preeclamptic pregnancies. Chen et al158 demonstrated the presence of an increased TNF- messenger ribonucleic acid expression in leukocytes from women with preeclampsia as compared with women with normal pregnancies and nonpregnant women. This high expression of TNF- may be associated with the TNF-1 allele, whose frequency is markedly increased in preeclampsia.158Opinion on whether placental antigens or the fetal genetic component is more relevant in the etiology of preeclampsia may be moving toward genetic (rather than immunologic) interpretation. According to Haig,224 a genetic conflict can be said to exist between maternal and fetal genes. Maternal and fetal genomes perform different roles during development. Heritable paternal, rather than maternal, imprinting of the genome is necessary for normal development of trophoblast and extraembryonic membranes.224 This is clearly revealed in aberrant development of human conceptuses with abnormal ratios of maternal-to-paternal genomes (a complete hydatidiform mole is a good example). Most complete moles are diploid (with both sets of chromosomes paternally derived) and are associated with a relatively high incidence of severe early-onset preeclampsia. Principally, parental imprinting refers to a parent-of-origin dependent expression of a subset of autosomal loci, independent of the sex of the offspring. McCowan and Becroft225 reported on the association of Beckwith-Wiedemann syndrome and preeclampsia. Overexpression of the IGF-2 gene during development may be the cause of excessive growth associated with the Beckwith-Wiedemann syndrome. It is interesting that the IGF-2 gene is subject to parental imprinting in humans.226 Animal experiments suggest that gene imprinting is also involved in placental HLA expression.227According to the genetic conflict theory, fetal genes will be selected to increase nutrient transfer to the fetus, and maternal genes will be selected to limit transfers in excess of some maternal optimum. The phenomenon of genomic imprinting means that a similar conflict exists within fetal cells between genes that are maternally derived and genes that are paternally derived. Haig224 stresses the fact that endovascular trophoblast invasion has 3 consequences: (1) The fetus gains direct access to its mothers arterial blood. Therefore a mother cannot reduce the nutrient content of blood reaching the placenta without reducing the nutrient supply to her own tissues. (2) The volume of blood reaching the placenta becomes largely independent of control by the local maternal vasculature. (3) The placenta is able to release hormones and other substances directly into the maternal circulation. The conflict hypothesis predicts that placental factors (fetal genes) will act to increase maternal blood pressure, whereas maternal factors will act to reduce blood pressure. It also suggests that the mother reduces vascular resistance early in pregnancy to ration fetal nutrients and that subsequent physiologic increase in vascular resistance represents the changing balance of power as the fetus grows larger. A corollary is that placental factors contribute to an increase in maternal cardiac output. Placental factors have an opportunity to preferentially increase nonplacental resistance because uteroplacental arteries are highly modified and unresponsive to vasoconstrictors.224 Intrinsic effects of a high maternal systemic blood pressure are ultimately beneficial to the fetus. Thus Haigs conflict hypothesis224 predicts that fetal genes will enhance maternal blood flow through the intervillous space by increasing maternal blood pressure (perfusion pressure). Maternal blood pressures form a continuum so that the dividing line between normotensive and hypertensive pregnancies is arbitrary. The conflict hypothesis predicts that a mothers position on this continuum is determined by the balance between fetal factors increasing blood pressure and maternal factors decreasing blood pressure. This mechanism may be operative in gestational hypertension, in which fetal prognosis is known to be good.224 In contrast, in preeclampsia the hypertension results from vasoconstriction rather than increased cardiac output. A hypoxic placenta may release cytotoxic factors that damage maternal endothelial cells98-102 and thus cause vasoconstriction and an increase in maternal blood pressure. In this situation small increments in the birth weight of semistarved fetuses may often have caused major increases in subsequent survival despite substantial costs to the mother. Thus endothelial cell dysfunction may have evolved as a high-risk fetal strategy to increase nonplacental resistance when the uteroplacental blood supply of a fetus is inadequate.

Underlying disorders associated with preeclampsiaTOP

Chronic hypertension, diabetes, and obesity predispose women to the development of preeclampsia.228-230 The link between obesity and preeclampsia is complex; many mechanisms may be involved. Obesity is characterized by expanded blood volume, increased cardiac output, and hypertension. Elevated cardiac output with compensatory vasodilatation may expose capillary beds to systemic pressures and increased flow, eventually leading to endothelial cell dysfunction characteristic of preeclampsia.231 Obesity is also associated with increased lipid availability, increased delivery of FFA to tissues, higher cholesterol and triglyceride levels, insulin resistance, and hyperinsulinemia.232,233 Women with preeclampsia are relatively insulin resistant and hyperinsulinemic during pregnancy, as well as months after delivery.234 These metabolic changes may, at least theoretically, cause a derangement of the VLDL/toxicity-preventing activity balance. High insulin levels may also directly cause hypertension, by increased sympathetic activity or increased tubular sodium resorption.235,236 High insulin levels in obesity may also aggravate cytokine-mediated oxidative stress.237 The combination of insulin resistance and compensatory hyperinsulinemia may be part of a cluster of abnormalities, the so-called syndrome X, including some degree of glucose intolerance, increased triglyceride levels, decreased high-density lipoprotein cholesterol levels, hypertension, hyperuricemia, smaller, denser low-density lipoprotein particles, and higher circulating levels of plasminogen activator inhibitor-1, which is more frequent in adults who themselves had growth restruction at birth.233,236Severe early-onset preeclampsia is also associated with a high incidence of underlying disorders such as protein S deficiency, activated protein C resistance [factor V mutation], hyperhomocysteinemia, and anticardiolipin antibodies, which may cause a more aggressive accelerated course of the disease by inducing an abnormal interaction between endothelial cells, platelets, leukocytes, and plasmatic coagulation and fibrinolytic factors.228 Unifying hypothesis (Fig 1)TOP

Fig. 1. Hypothesis on genesis of preeclampsia, in which most previously discussed concepts are integrated. EC, Endothelial cell; CTB, cytotrophoblast; STB, syncytiotrophoblast; CK, cytokine.

Click on Image to view full size

Endothelial cell dysfunction appears to be the final common pathway in the pathogenesis of preeclampsia, yet the etiology of the disease remains obscure. Genetic factors are involved; however, it is likely that there is not 1 major preeclampsia gene but several genetic factors that are associated with maternal susceptibility. Placental ischemia and increased syncytiotrophoblast deportation probably represent an end-stage disease phenomenon, actually very similar to that predicted by the conflict hypothesis, in which it is stated that endothelial cell damage may have evolved as a high-risk fetal strategy to increase nonplacental resistance when the uteroplacental blood supply to a fetus is inadequate.224 Immune maladaptation and the genetic conflict hypothesis are the hypotheses that are congruous with data on the effects of changing partners and the protective effect of sperm exposure. Presence of an abundance of decidual leukocytes and the complex cytokine network underlines the pivotal impact of a correct fetal (paternal)-maternal interaction. Cytokine-mediated oxidative stress, caused by immune maladaptation or a genetic conflict, fits with data on cytokines, oxidative stress, lipid changes, and the type and chronologic sequence of endothelial cell dysfunction in preeclampsia.

References TOP

1. 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;93:1049-59.

MEDLINE

2. Pijnenborg R, Anthony J, Davey DA, Rees A, Tiltman A, Vercruysse L, et al. Placental bed spiral arteries in the hypertensive disorders of pregnancy. Br J Obstet Gynaecol 1991;98:648-55.

MEDLINE

3. Pijnenborg R. The placental bed. Hypertens Pregnancy 1996;15:7-23.

4. Zhou Y, Damsky CH, Fisher SJ. Preeclampsia is associated with failure of human cytotrophoblasts to mimic a vascular adhesion phenotype. J Clin Invest 1997;99:2152-64.

MEDLINE

5. Meuris S, Nagy AM, Delogne-Desnoeck J, Jurkovic D, Jauniaux E. Temporal relationship between the human chorionic gonadotropin peak and the establishment of intervillous blood flow in early pregnancy. Hum Reprod 1995;10:947-50.

MEDLINE

6. Jaffe R, Dorgan A, Abramowicz JS. Color Doppler imaging of the uteroplacental circulation in the first trimester: value in predicting pregnancy failure or complication. Am J Roentgenol 1995;164:1255-8.

7. Meekins JW, Pijnenborg R, Hanssens M, McFadyen IR, van Assche A. A study of placental bed spiral arteries and trophoblast invasion in normal and severe pre-eclamptic pregnancies. Br J Obstet Gynaecol 1994;101:669-74.

MEDLINE

8. Meekins JW, Pijnenborg R, Hanssens M, van Assche A, McFadyen IR. Immunohistochemical detection of lipoprotein(a) in the wall of placental bed spiral arteries in normal and severe pre-eclamptic pregnancies. Placenta 1994;15:511-24.

MEDLINE

9. Charnock Jones DS, Sharkey AM, Boocock CA, Ahmed A, Plevin R, Ferrara N, et al. Vascular endothelial growth factor receptor localization and activation in human trophoblast and choriocarcinoma cells. Biol Reprod 1994;51:524-30.

10. Wheeler T, Elcock CL, Anthony FW. Angiogenesis and the placental environment. Placenta 1995;16:289-96.

MEDLINE

11. Rodesch F, Simon P, Donner C, Jauniaux E. Oxygen measurements in endometrial and trophoblastic tissues during early pregnancy. Obstet Gynecol 1992;80:283-5.

MEDLINE

12. Jauniaux E, Jurkovic D, Campbell S, Hustin J. Doppler ultrasonographic features of the developing placental circulation: correlation with anatomic findings. Am J Obstet Gynecol 1992;166:585-7.

MEDLINE

13. Hofmann GE, Glatsein I, Schatz F, Heller D, Deligdisch L. Immunohistochemical localization of urokinase-type plasminogen activator and the plasminogen activator inhibitors 1 and 2 in early human implantation sites. Am J Obstet Gynecol 1994;170:671-6.

MEDLINE

ABSTRACT

FULL TEXT

14. Moll UM, Lane BL. Proteolytic activity of first trimester human placenta: localization of interstitial collagenase in villous and extravillous trophoblast. Histochemistry 1990;94:555-60.

MEDLINE

15. Librach CL, Feigenbaum SL, Bass KE, Cui TY, Verastas N, Sadovsky Y, et al. Interleukin-l regulates human cytotrophoblast metalloproteinase activity and invasion in vitro. J Biol Chem 1994;269:17125-31.

MEDLINE

16. Hay ED. Collagen and other matrix glycoproteins in embryogen-esis. In: Hay ED, editor. Cell biology of extracellular matrix. New York: Plenum; 1991. p. 419-62.

17. Lin CQ, Bisell MJ. Multifaceted regulation of cell differentiation by extracellular matrix. FASEB J 1993;7:737-4.

MEDLINE

18. Graham CH, Lala PK. Mechanism of control of trophoblast invasion in situ. J Cell Physiol 1991;148:228-34.

MEDLINE

19. Aplin JD, Charlton AK, Ayad S. An immunohistochemical study of human endometrial extracellular matrix during the menstrual cycle and first trimester of pregnancy. Cell Tissue Res 1988;253:231-40.

MEDLINE

20. Hynes RO. The impact of molecular biology on models for cell adhesion. BioEssays 1994;16:663-9.

MEDLINE

21. Coutifaris C, Kao L, Sehdev HM, Chin U, Babalola G. E-Cadherin expression during the differentiation of trophoblasts. Development 1991;113:767-7.

MEDLINE

22. Loke YW, King A. Trophoblast interaction with extracellular matrix. In: Loke YW, King A, editors. Human implantation: cell biology and immunology. New York: Cambridge University Press; 1995. p. 151-79.

23. Hynes RO. Integrinsversatility, modulation, and signaling in cell adhesion. Cell 1992;69:11-25.

MEDLINE

24. Aplin JD. Expression of integrin 6b4 in human trophoblast and its loss from extravillous cells. Placenta 1993;14:203-15.

MEDLINE

25. Damsky CH, FitzGerald ML, Fisher SJ. Distribution patterns of extracellular matrix components and adhesion receptors are intricately modulated during first trimester cytotrophoblast differentiation along the invasive pathway, in vivo. J Clin Invest 1992;89:210-22.

MEDLINE

26. Burrows TD, King A, Loke YW. Expression of integrins by human trophoblast and differential adhesion to laminin or fibronectin. Hum Reprod 1993;8:475-84.

MEDLINE

27. Burrows TD, King A, Loke YW. Expression of adhesion molecules by endovascular trophoblast and decidual endothelial cells: implications for vascular invasion during implantation. Placenta 1994;15:21-33.

MEDLINE

28. Damsky CH, Librach C, Lim KH, FitzGerald ML, McMaster MT, Janatpour M, et al. Integrin switching regulates normal trophoblast invasion. Development 1994;120:3657-66.

MEDLINE

29. Zhou Y, Damsky CH, Chiu K, Roberts JM, Fisher SJ. Preeclampsia is associated with abnormal expression of adhesion molecules by invasive cytotrophoblasts. J Clin Invest 1993;91:950-60.

MEDLINE

30. Fisher SJ, Damsky CH. Human cytotrophoblast invasion. Semin Cell Biol 1993;4:183-8.

MEDLINE

31. Loke YW, King A. Clinical implications of defective implantation. In: Loke YW, King A, editors. Human implantation: cell biology and immunology. New York: Cambridge University Press, 1995; p. 224-36.

32. Pijnenborg R, Luyten C, Vercruysse L, Van Assche FA. Attachment and differentiation in vitro of trophoblast from normal and preeclamptic human placentas. Am J Obstet Gynecol 1996;175:30-6.

MEDLINE

ABSTRACT

FULL TEXT

33. Divers MJ, Bulmer JN, Miller D, Lilford RJ. Beta 1 integrins in third trimester human placentae: no differential expression in pathological pregnancies. Placenta 1995;16:245-60.

MEDLINE

34. Zhou Y, Chiu K, Brescia RJ, Combs CA, Katz MA, Kitzmiller JL, et al. Increased depth of trophoblast invasion after chronic constriction of the lower aorta in rhesus monkeys. Am J Obstet Gynecol 1993;169:224-9.

MEDLINE

35. Meagher EA, FitzGerald GA. Disordered eicosanoid formation in pregnancy-induced hypertension. Circulation 1993;88;1324-33.

MEDLINE

36. Wang Y, Walsh SW, Guo J, Zhang J. Maternal levels of prostacyclin, thromboxane, vitamin E and lipid peroxides throughout normal pregnancy. Am J Obstet Gynecol 1991;165:1690-4.

MEDLINE

37. Gant NF, Daley GI, Chand S, Whalley PJ, MacDonald PC. A study of angiotensin-II pressor response throughout primigravid pregnancy. J Clin Invest 1973;52:2682-9.

MEDLINE

38. Conrad KP, Colpoys MC. Evidence against the hypothesis that prostaglandins are the vasodepressor agents of pregnancy. J Clin Invest 1986;77:236-45.

MEDLINE

39. Mukaddam Daher S, Moutquin JM, Gutkowska BJ, Nuwayhid BS, Quillen EW Jr. Effects of prostaglandin inhibition on the renal function curve during ovine pregnancy. Am J Obstet Gynecol 1994;171:599-607.

MEDLINE

ABSTRACT

FULL TEXT

40. Fay TN, Grudzinskas JG. Human endometrial peptides: a review of their potential role in implantation and placentation. Hum Reprod 1991;6:1311-26.

MEDLINE

41. Daunter B. Immunology of pregnancy: towards a unifying hypothesis. Eur J Obstet Gynecol Reprod Biol 1992;43:81-95.

MEDLINE

42. Morris NH, Eaton BM, Dekker G. Nitric oxide, the endothelium, pregnancy and pre-eclampsia. Br J Obstet Gynaecol 1996;103:4-15.

MEDLINE

43. van Buren GA, Yang D, Clark KE. Estrogen-induced uterine vasodilatation is antagonized by L-nitroarginine methyl ester, an inhibitor of nitric oxide synthesis. Am J Obstet Gynecol 1992;167:828-33.

MEDLINE

44. Weiner CP, Lizasoain I, Baylis SA, Knowles RG, Charles IG, Moncada S. Induction of calcium-dependent nitric oxide synthases by sex hormones. Proc Natl Acad Sci U S A 1994;91: 5212-6.

MEDLINE

45. Conrad KP, Joffe GM, Kruszyna H, Kruszyna R, Rochntitryp P, Chavez JE, et al. Identification of increased nitric oxide biosynthesis during pregnancy in rats. FASEB J 1993;7:566-71.

MEDLINE

46. Chu ZM, Beilin LJ. Nitric oxidemediated changes in vascular reactivity in pregnancy in spontaneously hypertensive rats. Br J Pharmacol 1993;110:1184-8.

MEDLINE

47. Ahokas RA, Mercer BM, Sibai BM. Enhanced endothelium-derived relaxing factor activity in pregnant spontaneously hypertensive rats. Am J Obstet Gynecol 1991;165:801-7.

MEDLINE

48. Molnar M, Suto T, Toth 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;170:1458-66.

MEDLINE

ABSTRACT

FULL TEXT

49. Pascoal IF, Umans JG. Effect of pregnancy on mechanisms of relaxation in human omental microvessels. Hypertension 1996;28:183-7.

MEDLINE

50. Myatt L, Brewer A, Langdon G, Brockman DE. Attenuation of the vasoconstrictor effects of thromboxane and endothelin by nitric oxide in the human fetal-placental circulation. Am J Obstet Gynecol 1992;166:224-30.

MEDLINE

51. Myatt L, Brockman DE, Langdon G, Pollock JS. Constitutive calcium-dependent isoform of nitric oxide synthase in the human placental villous vascular tree. Placenta 1993;14:373-83.

MEDLINE

52. Myatt L, Brockman DE, Eis AL, Pollock JS. Immunohistochemical localization of nitric oxide synthase in the human placenta. Placenta 1993;14:487-95.

MEDLINE

53. Buttery LDK, McCarthy A, Springall DR, Sullivan MHF, Elder MG, Michel T, et al. Endothelial nitric oxide synthase in the human placenta: regional distribution and proposed regulatory role at the fetomaternal interface. Placenta 1994;15:257-65.

MEDLINE

54. Morris NH, Sooranna SR, Learmont JG, Poston L, Ramsey B, Pearson JD, et al. Nitric oxide synthase activities in placental tissue from normotensive, preeclamptic and growth retarded pregnancies. Br J Obstet Gynaecol 1995;102:711-4.

MEDLINE

55. Toth M, Kukor Z, Romero R, Hertelendy F. Nitric oxide synthase in first-trimester human placenta: characterization and subcellular distribution. Hypertens Preg 1995;14:287-300.

56. Baker PN, Broughton Pipkin F, Symonds EM. Platelet angiotensin II binding sites in normotensive and hypertensive women. Br J Obstet Gynaecol 1991;98:436-40.

MEDLINE

57. Friedman SA. Preeclampsia: a review of the role of prostaglandins. Obstet Gynecol 1988;71:122-37.

MEDLINE

58. FitzGerald DJ, Rocki W, Murray R, Mayo G, FitzGerald GA. Thromboxane A2 synthesis in pregnancy-induced hypertension. Lancet 1990;335:751-4.

MEDLINE

59. Walsh SW. Preeclampsia: an imbalance in placental prostacyclin and thromboxane production. Am J Obstet Gynecol 1985;152:335-40.

MEDLINE

60. Woodworth SH, Li X, Lei ZM, Rao CV, Yussman MA, Spinnato JA II, et al. Eicosanoid biosynthetic enzymes in placental and decidual tissues from preeclamptic pregnancies: increased expression of thromboxane-A2 synthase gene. J Clin Endocrinol Metab 1994;78:1225-31.

MEDLINE

61. Zeeman GG, Dekker GA. Pathogenesis of preeclampsia: a hypothesis. Clin Obstet Gynecol 1992;35:317-37.

MEDLINE

62. Friedman SA, de Groot CJM, Taylor RN, Golditch BD, Roberts JM. Plasma cellular fibronectin as a measure of endothelial involvement in preeclampsia and intrauterine growth retardation. Am J Obstet Gynecol 1994;170:838-41.

MEDLINE

ABSTRACT

FULL TEXT

63. Hsu CD, Iriye B, Johnson TRB, Witter FR, Hong SF, Chan DW. Elevated circulating thrombomodulin in severe preeclampsia. Am J Obstet Gynecol 1993;169:148-9.

MEDLINE

64. Friedman SA, Schiff E, Emeis JJ, Dekker GA, Sibai BM. Biochemical corroboration of endothelial involvement in severe preeclampsia. Am J Obstet Gynecol 1995;172:202-3.

MEDLINE

ABSTRACT

FULL TEXT

65. de Boer K, Lecander I, ten Cate JW, Borm JJJ, Treffers PE. Placental-type plasminogen activator inhibitor in preeclampsia. Am J Obstet Gynecol 1988;158:518-22.

MEDLINE

66. de Groot CJM, Taylor RN. The role of endothelium in pregnancy and pregnancy-related diseases. Curr Obstet Med 1993;2:107-37.

67. Roberts JM, Taylor RN, Goldfien A. Endothelial cell activation as a pathogenetic factor in preeclampsia. Semin Perinatol 1991;15:86-93.

MEDLINE

68. Roberts JM, Taylor RN, Goldfien A. Clinical and biochemical evidence of endothelial cell dysfunction in the pregnancy syndrome preeclampsia. Am J Hypertens 1991;4:700-8.

MEDLINE

69. Kamoi K, Sudo N, Ishibashi M, Yamaji T. Plasma endothelin-1 levels in patients with pregnancy-induced hypertension. N Engl J Med 1990;323:1486-7.

MEDLINE

70. Nova A, Sibai BM, Barton JR, Mercer BM, Mitchell MD. Maternal plasma level of endothelin is increased in preeclampsia. Am J Obstet Gynecol 1991;165:724-7.

MEDLINE

71. Taylor RN, Varma M, Teng NH, Roberts JM. Women with preeclampsia have higher plasma endothelin levels than women with normal pregnancies. J Clin Endocrinol Metab 1990;71:1675-7.

MEDLINE

72. Kraayenbrink AA, Dekker GA, van Kamp GJ, van Geijn HP. Endothelial vasoactive mediators in preeclampsia. Am J Obstet Gynecol 1993;169:160-5.

MEDLINE

73. Barton JR, Sibai BM, Whybrew WD, Mercer BM. Urinary endothelin-l: not a useful marker for preeclampsia. Am J Obstet Gynecol 1993;168:599-601.

MEDLINE

74. Shanklin DR, Sibai BM. Ultrastructural aspects of preeclampsia. I. Placental bed and uterine boundary vessels. Am J Obstet Gynecol 1989;161:735-41.

MEDLINE

75. Shanklin DR, Sibai BM. Ultrastructural aspects of preeclampsia. II. Mitochondrial changes. Am J Obstet Gynecol 1990;163:943-53.

MEDLINE

76. Barton JR, Hiett AK, OConnor WN, Nissen SE, Greene JW. Endomyocardial ultrastructural findings in preeclampsia. Am J Obstet Gynecol 1991;165:389-91.

MEDLINE

77. Endresen MJ, Tosti E, Lorentzen B, Henriksen T. Sera of preeclamptic women are not cytotoxic to endothelial cells in culture. Am J Obstet Gynecol 1995;172:196-201.

MEDLINE

ABSTRACT

FULL TEXT

78. McCarthy AL, Woolfson RG, Raju SK, Poston L. Abnormal endothelial cell function of resistance arteries from women with preeclampsia. Am J Obstet Gynecol 1993;168:1323-30.

MEDLINE

79. Friedman SA, Lubarsky SL, Ahokas RA, Nova A, Sibai BM. Preeclampsia and related disorders: clinical aspects and relevance of endothelin and nitric oxide. Clin Perinatol 1995;22:343-55.

MEDLINE

80. Davidge ST, Billiar TR, Roberts JM. Plasma nitrites and nitrates, stable end products of nitric oxide, are elevated in women with preeclampsia. Hypertens Pregn 1994;13:344.

81. Cameron IT, van Papendorp CL. Relationship between nitric oxide synthesis and increase in systolic blood pressure in women with hypertension in pregnancy. Hypertens Pregn 1993;12:85-92.

82. Curtis NE, Gude NM, King RG, Marriott PJ, Rook TJ, Brennecke SP. Nitric oxide in normal human pregnancy and preeclampsia. Hypertens Pregn 1995;14:339-49.

83. Lyall F, Young A, Greer IA. Nitric oxide concentrations are increased in the fetoplacental circulation in preeclampsia. Am J Obstet Gynecol 1995;173:714-8.

MEDLINE

ABSTRACT

FULL TEXT

84. Brown MA, Tibben E, Zammit VC, Cario GM, Carlton MA. Nitric oxide excretion in normal and hypertensive pregnancies. Hypertens Pregn 1995;14:319-26.

85. Davidge ST, Stranko CP, Roberts JM. Urine but not plasma nitric oxide metabolites are decreased in women with preeclampsia. Am J Obstet Gynecol 1996;174:1008-13.

MEDLINE

ABSTRACT

FULL TEXT

86. Vallance P, Collier J. Biology and clinical relevance of nitric oxide. BMJ 1994;309:453-7.

MEDLINE

87. Fickling SA, Williams D, Vallance P, Nussey SS, Whitley GS. Plasma concentrations of endogenous inhibitor of nitric oxide synthesis in normal pregnancy and pre-eclampsia. Lancet 1993;342:242-3.

MEDLINE

88. Ramsay B, de Belder A, Campbell S, Moncada S, Martin JF. A nitric oxide donor improves uterine artery diastolic blood flow in normal early pregnancy and in women at high risk of pre-eclampsia. Eur J Clin Invest 1994;24:76-8.

MEDLINE

89. Ballegeer VC, Spitz B, De Baene LA, van Assche AF, Hidajar M, Criel AM. Platelet activation and vascular damage in gestational hypertension. Am J Obstet Gynecol 1992;166:629-33.

MEDLINE

90. Halligan A, Bonnar J, Sheppard B, Darling M, Walshe J. Haemostatic, fibrinolytic and endothelial variables in normal and pre-eclamptic pregnancies. Br J Obstet Gynaecol 1994;101:488-92.

MEDLINE

91. Middelkoop CM, Dekker GA, Kraayenbrink AA, Popp-Snijders C. Platelet-poor plasma serotonin in normal and preeclamptic pregnancy. Clin Chem 1993;39:1675-8.

MEDLINE

92. Vanhoutte PM. Serotonergic antagonists and vascular disease. Cardiovasc Drugs Ther 1990;4:7-12.

MEDLINE

93. de Jong CLD, Dekker GA, Sibai BM. The renin-angiotensin-aldosterone system in preeclampsia: a review. Clin Perinatol 1991;18:1-29.

94. Mitsuhashi N, Kuwabara Y. Prostaglandin biosynthesis in the umbilical artery and its relation to fetal growth. Asia Oceania J Obstet Gynaecol 1994;20:111-4.

MEDLINE

95. Wang Y, Walsh SW, Parnell R, Han J. Placental production of nitric oxide and endothelin in normal and preeclamptic pregnancies. Hypertens Pregn 1994;13:171-8.

96. Ghabour MS, Eis ALW, Brockman DE, Pollock JS, Myatt L. Immunohistochemical characterization of placental nitric oxide synthase expression in preeclampsia. Am J Obstet Gynecol 1995;173:687-94.

MEDLINE

ABSTRACT

FULL TEXT

97. Eis ALW, Brockman DE, Pollock JS, Myatt L. Immunohistochemical localization of endothelial nitric oxide synthase in human villous and extravillous trophoblast populations and expression during syncytiotrophoblast formation in vitro. Placenta 1995;16:113-26.

MEDLINE

98. Smarason AK, Sargent IL, Starkey PM, Redman CWG. The effect of placental syncytiotrophoblast microvillous membranes from normal and pre-eclamptic women on the growth of endothelial cells in vivo. Br J Obstet Gynaecol 1993;100:943-9.

MEDLINE

99. Sargent IL, Johansen M, Chua S, Redman CWG. Clinical experience: isolating trophoblasts from maternal blood. Ann N Y Acad Sci 1994;731:154-61.

MEDLINE

100. Chua S, Wilkins T, Sargent I, Redman C. Trophoblast deportation in pre-eclamptic pregnancy. Br J Obstet Gynaecol 1991;98:973-9.

MEDLINE

101. Arkwright PD, Rademacher TW, Dwek RA, Redman CWG. Preeclampsia is associated with an increase in trophoblast glycogen content and glycogen synthase activity, similar to that found in hydatidiform moles. J Clin Invest 1993;91:2744-53.

MEDLINE

102. Ding Z, Rowe J, Sinosich M, Saunders DM, Gallery EDM. Secretion of vasoactive substances by cultured human villous cytotrophoblasts. Hypertens Pregn 1994;13:352-7.

103. Tsukimori K, Ishida K, Maeda H, Koyanagi T, Nakano H. The placenta as possible source of the factor causing endothelial cell injury in preeclampsia. Hypertens Pregn 1994;13:358.

104. Lockwood CJ, Peters JH. Increased plasma levels of ED1(+) cellular fibronectin precede the clinical signs of preeclampsia. Am J Obstet Gynecol 1990;162:358-62.

MEDLINE

105. Taylor RN, Heilbron DC, Roberts JM. Growth factor activity in the blood of women in whom preeclampsia develops is elevated from early pregnancy. Am J Obstet Gynecol 1990;163:1839-44.

MEDLINE

106. Cross JC. Trophoblast function in normal and preeclamptic pregnancy. Fetal Matern Med Rev 1996;8:57-66.

107. Lorentzen B, Endresen MJ, Clausen T, Henriksen T. Fasting serum free fatty acids and triglycerides are increased before 20 weeks of gestation in women who later develop preeclampsia. Hypertens Pregn 1994;13:103-9.

108. Endresen MJ, Lorentzen B, Henriksen T. Increased lipolytic activity and high ratio of free fatty acids to albumin in sera from women with preeclampsia leads to triglyceride accumulation in cultured endothelial cells. Am J Obstet Gynecol 1992;167:440-7.

MEDLINE

109. Endresen MJ, Lorentzen B, Henriksen T. Increased lipolytic activity of sera from pre-eclamptic women due to the presence of a lysophospholipase. Scand J Clin Lab Invest 1993;53:733-9.

MEDLINE

110. Endresen MJ, Tosti E, Heimli H, Lorentzen B, Henriksen T. Effects of free fatty acids found increased in women who develop pre-eclampsia on the ability of endothelial cells to produce prostacyclin, cGMP and inhibit platelet aggregation. Scand J Clin Lab Invest 1994;54:549-57.

MEDLINE

111. Arbogast BW, Leeper SC, Merrick RD, Olive KE, Taylor RN. Hypothesis: which plasma factors bring about disturbance of endothelial function in pre-eclampsia? Lancet 1994;343:340-1.

MEDLINE

112. Lorentzen B, Drevon CA, Endresen MJ, Henriksen T. Fatty acid pattern of esterified and free fatty acids in sera of women with normal and pre-eclamptic pregnancy. Br J Obstet Gynaecol 1995;102:530-7.

MEDLINE

113. Sattar N, Gaw A, Packard CJ, Greer IA. Potential pathogenic roles of aberrant lipoprotein and fatty acid metabolism in preeclampsia. Br J Obstet Gynaecol 1996;103:614-20.

MEDLINE

114. Loke YW, King A. Immunology of human placental implantation: clinical implications of our current understanding. Mol Med Today 1997;Apr:153-9.

115. Sargent IL. Maternal and fetal immune responses during pregnancy. Exp Clin Immunogenet 1993;10:85-102.

MEDLINE

116. Carosella ED, Dausset J, Kirszenbaum M. HLA-G revisited. Immunol Today 1996;17:407-9.

MEDLINE

117. Shorter SC, Starkey PM, Ferry BL, Clover LM, Sargent IL, Redman CWG. Antigenic heterogeneity of human cytotrophoblast and evidence for the transient expression of MHC class I antigens distinct from HLA-G. Placenta 1993;14:571-82.

MEDLINE

118. van der Ven K, Ober C. HLA-G polymorphisms in African Americans. J Immunol 1994;153:5628-33.

MEDLINE

119. Starkey PM. The decidua and factors controlling placentation. In: Redman CWG, Sargent LL, Starkey PM, editors. The human placenta. Oxford (United Kingdom): Blackwell Scientific Publications; 1993. p. 362-413.

120. Saito S, Nishikawa K, Morii T, Enomoto M, Narita N, Motoyoshi K, et al. Cytokine production by CD16 CD56 bright natural killer cells in human pregnancy decidua. Int Immunol 1993;5:559-63.

MEDLINE

121. Deniz G, Christmas SE, Brew R, Johnson PM. Phenotypic and functional cellular differences between human CD3(-) decidual and peripheral blood leukocytes. J Immunol 1994;152:4255-61.

MEDLINE

122. Klein J, Ohuigin C. The conundrum of nonclassical major histocompatibility complex genes. Proc Natl Acad Sci U S A 1994;91:6251-2.

MEDLINE

123. Chumbley G, King A, Robertson K, Holmes N, Loke YW. Resistance of HLA-G and HLA-A2 transfectants to lysis by decidual NK cells. Cell Immunol 1994;155:312-22.

MEDLINE

124. Ljunggren HG, Karre K. In search of the missing self MHC molecules and NK cell recognition. Immunol Today 1990;11:237-44.

MEDLINE

125. Jurisicova A, Casper RF, MacLusky NJ, Mills GB, Librach CL. HLA-G expression during preimplantation human embryo development. Proc Natl Acad Sci U S A 1996;93:161-5.

MEDLINE

126. McMaster MT, Librach CL, Zhou Y, Lim KH, Janatpour MJ, DeMars R, et al. Human placental HLA-G expression is restricted to differentiated cytotrophoblasts. J Immunol 1995;154:3771-8.

MEDLINE

127. Colbern GT, Chiang MH, Main EK. Expression of the nonclassic histocompatibility antigen HLA-G by preeclamptic placenta. Am J Obstet Gynecol 1994;170:1244-50.

MEDLINE

ABSTRACT

FULL TEXT

128. King A, Balendran N, Wooding P, Carter NP, Loke YW. CD3 leukocytes present in the human uterus during early placentation: phenotypic and morphologic characterization of the CD56++ population. Dev Immunol 1991;1:169-90.

MEDLINE

129. Wegmann TG. Maternal T cells promote placental growth and prevent spontaneous abortion. Immunol Lett 1988;17:297-302.

MEDLINE

130. Shorter SC, Vince GS, Starkey PM. Production of granulocyte-colony stimulating factor at the materno-foetal interface in human pregnancy. Immunology 1992;75:468-74.

MEDLINE

131. Giudice LC. Growth factors and growth modulators in human uterine endometrium: their potential relevance to reproductive medicine. Fertil Steril 1994;61:1-17.

MEDLINE

132. Aoki K, Kajiura S, Matsumoto Y, Ogasawara M, Okada S, Yagami Y, et al. Preconceptional natural-killer-cell activity as a predictor of miscarriage. Lancet 1995;345:1340-2.

MEDLINE

133. Clark DA. Cytokines, decidua, and early pregnancy. Oxf Rev Reprod Biol 1993;15:83-111.

MEDLINE

134. Giudice LC. Growth factors and growth modulators in human uterine endometrium: their potential relevance to reproductive medicine. Fertil Steril 1994;61:1-17.

MEDLINE

135. Mosmann TR, Cherwinski H, Bond MW. Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins. J Immunol 1986;136:2348-57.

MEDLINE

136. Mosmann TR, Moore KW. The role of Il-10 in cross-regulation of Thl and Th2 responses. Immunol Today 1991;12:A49-53.

MEDLINE

137. Wegmann TG, Lin H, Guilbert L, Mosmann TR. Bidirectional cytokine interactions in the maternal-fetal relationship: is successful pregnancy a Th2 phenomenon? Immunol Today 1993;15:353-6.

138. Varner MW. Autoimmune disorders and pregnancy. Semin Perinatol 1991;15:238-50.

MEDLINE

139. Cadet P, Rady PL, Tyring SK, Yandell RB, Hughes TK. Interleukin-10 messenger ribonucleic acid in human placenta: implications of a role for interleukin-10 in fetal allograft protection. Am J Obstet Gynecol 1995;173:25-9.

MEDLINE

ABSTRACT

FULL TEXT

140. Goodwin JS, Ceuppens J. Regulation of the immune response by prostaglandins. J Clin Immunol 1983;3:295-315.

MEDLINE

141. Lin H, Mosman