See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/49735733

Etiology of preeclampsia: an update

Article in Journal of the Medical Association of Thailand = Chotmaihet thangphaet · October 2004
Source: PubMed

CITATIONS READS

7 268

2 authors, including:

Gustaaf A Dekker
University of Adelaide
418 PUBLICATIONS 14,797 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Thesis on Psychosocial Factors and Pregnancy Outcome; a biopsychosocial study View project

Comprehension of preeclampsia View project

All content following this page was uploaded by Gustaaf A Dekker on 21 July 2015.

The user has requested enhancement of the downloaded file.

 Etiology of Preeclampsia: An Update
Gus Dekker *,
Nares Sukcharoen **

* Department of Obstetrics and Gynaecology, University of Adelaide, Lyell McEwin Health Service
** Department of Obstetrics and Gynaecology, Faculty of Medicine, Chulalongkorn University

Preeclampsia still ranks as one of obstetrics major problems. Clinicians typically encounter
preeclampsia as maternal disease with variable degrees of fetal involvement. More and more the unique
immunogenetic maternal - paternal relationship is appreciated, and as such also the specific ‘genetic
conflict’ that is characteristic of haemochorial placentation. Factors influencing the unique maternal-fetal
(paternal) interaction probably include the length and type of sexual relationship, the maternal (decidual
natural killer cells) acceptation of the invading cytotrophoblast (paternal HLA-C), and seminal levels of
transforming growth factor-b and probably other cytokines. The magnitude of the maternal response would
be determined by factors including a maternal set of genes determining her characteristic inflammatory
responsiveness, age, quality of her endothelium, obesity/ insulin resistance and probably a whole series of
susceptibility genes amongst which the thrombophilias received a lot of attention in recent years.

Keywords : Etiology, Preeclampsia

J Med Assoc Thai 2004; 87(Suppl 3): S96-103

Shallow, endovascular cytotrophoblast inva- early pathogenesis of preeclampsia are currently

sion in the spiral arteries, inappropriate endothelial popular(3). It should be stressed that these hypotheses
cell activation and an exaggerated inflammatory are certainly not mutually exclusive, but most likely
response are key features in the pathogenesis of pre- interactive to some extend:
eclampsia. In the late 1980s, generalized endothelial
dysfunction was considered to be one the triggering 1. The placental ischemia hypothesis
steps in its pathogenesis(1). More recently evidence Increased trophoblast deportation, as a
has been gathered that inappropriate endothelial acti- consequence of placental ischemia, may inflict endo-
vation is part of more generalized intravascular inflam- thelial cell dysfunction. In more recent publications,
matory reaction involving intravascular leucocytes as the Oxford investigators2 explain that poor placenta-
well as the clotting and complement systems(2). tion should be considered to be a separate pathologic
Preeclampsia occurs in 3-5% of pregnancies mechanism, not the cause of preeclampsia but rather a
and is a major cause (15-20%) of maternal mortality powerful predisposing factor. In the adapted form
in developed countries and a leading cause of this hypothesis dictates that poor placentation is a
(iatrogenic) preterm birth and intra-uterine growth separate disorder that once established usually but
restriction (IUGR). Understanding the etiology and not always leads to the maternal syndrome, depending
pathogenesis of preeclampsia would be the major on the extent to which it causes inflammatory signals
breakthrough in clinical obstetrics. This review (which may depend on fetal genes) and the nature of
aims to provide un update on our current level of the maternal response to those signals (which would
understanding of preeclampsia, still standing as the depend on maternal genes).
NUMBER ONE major obstetrical syndrome.
A number of hypotheses on the etiology and 2. The very low-density lipoprotein (VLDL) versus
Correspondence to : Dekker G, Department of Obstetrics and
toxicity-preventing activity (TxPA) hypothesis
Gynaecology, University of Adelaide, Lyell McEwin Health In preeclampsia, circulating free fatty acids
Service, Haydown Road, Elizabeth Vale, 5112 SA. (FFA) are increased already 15-20 weeks before the

S96 J Med Assoc Thai Vol. 87 Suppl. 3 2004

onset of disease. These FFA have a variety of adverse mother unable to cope with a ‘physiologic’ genetic
effects on endothelial physiology. Plasma albumin conflict.
exists as several isoelectric species, which range from
isoelectric point (pI) 4.8 to pI 5.6. The more FFA are Risk Factors
bound to albumin the lower the pI. Plasma albumin The etiology of preeclampsia is still
exerts a toxicity-preventing activity (TxPA) if it is in unknown. Many risk factors have been identified(7)
the pI 5.6 form. Since higher ratios of FFA to albumin (Table 1). Knowing these risks factors is useful for
cause a shift from the pI 5.6 to the pI 4.8 form of the clinician to target groups of patients at a higher
albumin, preeclamptic patients will have lower amounts risk of developing preeclampsia.
of protective TxPA (pI 5.6) than normotensive
pregnant women. A low ratio of TxPA to VLDL would Sperm Exposure, Primipaternity and the Paternal
result in cytotoxicity and triglyceride accumulation in Factor
endothelial cells. In some pregnant women (low Generally, preeclampsia occurs in first preg-
albumin concentrations?), the burden of transporting nancies(1). A previous normal pregnancy is associated
extra FFA from adipose tissue to the liver as a response with a markedly lowered incidence of preeclampsia,
to the increase energy demands, is likely to reduce even a previous abortion provides some protection
the concentration of TxPA to a point where VLDL in this respect(8). The protective effect of being multi-
toxicity is expressed, leading to endothelial injury(4,5). parous, is however lost with change of partner. The
term primipaternity was introduced by Robillard et al
3. The hyperdynamic disease model
According to the hyperdynamic disease
Table 1. Risk factors for preeclampsia
model, early in pregnancy, preeclamptic patients
have an elevated cardiac output with compensatory PRECONCEPTIONAL AND/OR CHRONIC RISK FACTORS
vasodilatation. The dilated systemic terminal arterioles Partner-related risk factors
and renal afferent arterioles may expose capillary nulliparity/primipaternity/teenage pregnancy
beds to syste-mic pressures and increased flow, even- limited sperm exposure, donor insemination, oocyte donation
oral sex (risk reduction)
tually leading to endothelial cell injury characteristic partner who fathered a preeclamptic pregnancy in another woman
of the ‘injury’ seen in preeclampsia(6).
Non-partner related risk factors
4. The immune maladaptation hypothesis history of previous preeclampsia
The interaction between decidual leukocytes age, interval between pregnancies
family history
and invading cytotrophoblast cells is essential for
normal trophoblast invasion and development. Immune Presence of specific underlying disorders
maladaptation may cause the shallow invasion of chronic hypertension and renal disease
spiral arteries by endovascular cytotrophoblast obesity, insulin resistance, low birth weight
cells and endothelial cell dysfunction mediated by gestational diabetes, type I diabetes mellitus
activated protein C resistance (factor V mutation), protein S defi-
an increased decidual release of Th1 cytokines, ciency
proteolytic enzymes, and free radical species(3). antiphospholipid antibodies
hyperhomocysteinemia
5. The genetic hypothesis sickle cell disease, sickle cell trait (?)
The development of preeclampsia-eclampsia
Exogenous factors
may be based on a single recessive gene or a dominant smoking (risk reduction)
gene with incomplete penetrance. Penetrance may be stress, work related psycho-social strain
dependent on the fetal genotype. in utero DES exposure

6. The genetic conflict hypothesis PREGNANCY-ASSOCIATED RISK FACTORS

multiple pregnancy
The maternal and fetal genomes perform structural congenital anomalies
different roles during development. Inheritable pater- hydrops fetalis
nal, rather than maternal, imprinting of the genome is chromosomal anomalies (trisomy 13, triploidy)
necessary for normal trophoblast development. Pre- hydatidiform moles
eclampsia may relate to a ‘hefty’ genetic conflict, or a urinary tract infection

J Med Assoc Thai Vol. 87 Suppl. 3 2004 S97

(9,10)
exploring the relationship between severe preec- the first pregnancy was 13.1% if she had her second
lampsia and changes in pater-nity patterns among pregnancy with the same partner. This risk dropped to
multig-ravidae. 11.8%, if she changed the partner19. If anything one
Change of partner in a subsequent pregnancy would assume a longer birth interval after experiencing a
(10,11)
or oocyte donation(12) increase a woman’s risk traumatic life event, such as a pregnancy complicated
of preeclampsia, suggesting that prior exposure to by preeclampsia. The results from these studies
fetal (paternal) antigens is protective(13). Robillard contradict the ‘birth interval hypothesis’(18,19).
postulated that preeclampsia may be a problem of As mentioned earlier, the duration of sexual
primipaternity rather than primigravidity(10). Previous cohabitation is also an important determining risk
sperm exposure also conveys protection. The risk of factor. The number of sexual cohabitation’s preceding
preeclampsia was tripled in women pregnant by pregnancy is about three times higher in normal
intracytoplasmic sperm injection (ICSI) done with pregnant women than in preeclamptic women and this
surgically obtained sperm (i.e. these women were finding provides an explanation for the high incidence
never exposed to their partner’s sperm) compared to of preeclampsia in teenage pregnant girls(20). In a case-
women pregnant after in vitro fertilization (IVF) or ICSI control study comparing the contraceptive histories
with ejaculated sperm14. Oral tolerization to paternal of primiparous women with and without preeclampsia,
antigens by oral sex and swallowing sperm is the risk of preeclampsia for users of contraceptives
associated with a lower incidence of preeclampsia(15). that prevent exposure to sperm increases 2.4 fold(21).
A recent Norwegian study16 claimed that it In a prospective study on the relationship between
is prolonged birth interval and not primipaternity as sperm exposure and preeclampsia,the incidence of
risk factor for preeclampsia in multiparous women. pregnancy-induced hypertension was 11.9% among
After adjustment for the interval between births in a primigravidae, 4.7% among samepaternity multi-
Norwegian nationwide study, change of partner was gravidae, and 24.0% among new-paternity multi-
no longer associated with an increased risk of gravidae. For both primigravidae and multigravidae,
preeclampsia. However, the latter study was based on length of sexual cohabitation before conception
data from the Norwegian Birth Registry. It is known was inversely related to the incidence of pregnancy-
that birth registries do not contain the details necessary induced hypertension (p < 0.0001). Taking women
to adequately investigate paternity in humans. There cohabitating for more than 12 months as reference, a
is a significant rate (5-20%) of false claimed paternities cohabitation period of 0-4 months was shown to be
in stable couples in developed countries(17). Those associated with a typical odds ratio (OR) of 11.6, a
nevertheless fall into the ‘same father’ group in the period of 5-8 months with a typical OR of 5.9, and a
latter study. Conclusions concerning the association period of 9-12 months with a typical OR of 4.2. The
between paternity and preeclampsia based on birth very high incidence (24.0%) of pregnancy-induced
registry alone should be avoided, since they are a hypertension among new-paternity multiparous
very crude indication of the real paternities and do women was found to be related to a remarkable short
not contain the necessary information on sexual period of sperm exposure preceding conception(10).
cohabitation. In this study, when the interbirth inter- The exact way by which the female organism
val in multiparous women was 10 years or more, the is exposed to the paternal HLA message is uncertain.
risk for preeclampsia approximated that among Sperm exposure does provide protection against
nulliparous women. This may actually have revealed developing preeclampsia. Actual exposure to the
genetic vascular/thrombophillic predisposition for sperm cells appears to be important for that matter.
preeclampsia which are well-known risk factors for Deposition of semen in the female genital tract
preeclampsia with advancing age, or may have included provokes a cascade of cellular and molecular events
women with one or more miscarriages between the that resemble a classic inflammatory response. The
births. They fail to discuss the findings of the critical seminal factor appears to be seminal vesicle
previous study showing that the effect of partner derived transforming growth factor 1 (TGF1). Seminal
change was influenced by the outcome of the preced- vesicle-derived TGF1 is secreted predominantly in a
ing pregnancy. Partner change was found to convey latent form. Seminal plasmin and uterine factors
protection if the first pregnancy was complicated by transform the latent form into bioactive TGF1(22). Intra-
preeclampsia18. Interestingly, the risk to develop uterine insemination of TGF1 in-vivo results in an
preeclampsia for mothers who had preeclampsia in increase in granulocyte-monocyte colony stimulating

S98 J Med Assoc Thai Vol. 87 Suppl. 3 2004

factor (GM-CSF) production that is sufficient to nancies; duplication and fragmentation of the internal
initiate an endometrial leukocytosis comparable with elastic lamina and the proportion of non-muscular
that seen following mating(22). The introduction of tissue increased with increasing parity(26). It would be
TGF1 into the uterus in combination with paternal very important to know whether or not these changes
ejaculate antigens favours the growth and survival of regress with prolonged birth interval.
the semi-allogenic fetus, as evidenced by a signifi- To reduce the risk of preeclampsia/eclampsia,
cant increase in fetal and placental weight in animal it is better for the human female to avoid conception
studies, in two ways. Firstly, by initiating a post- soon after initiating sexual relations with a new partner,
mating inflammatory reaction, TGF1 increases the regardless of her gravidity. The human female has to
ability to sample and process paternal antigens tolerate the semi-allogeneic graft through the father’s
contained within the ejaculate. Another important role alloantigens, and this is accomplished after an
of TGF1 and the subsequent post-coital inflammatory immunological stimulation following an extended
response is the initiation of a strong Type 2 immune duration of sexual exposure to the father’s sperm. In
deviation. The processing of an antigen by antigen- the case of a first pregnancy with a particular father,
presenting-cells in an environment containing TGF1 she is able to tolerate it given a long constant exposure
is likely to initiate a Th2 phenotype within these to the father’s antigens through his sperm(11,12,27).
responding T-cells(23). By initiating a Type 2 immune All these findings on the effects of partner change,
response towards paternal ejaculate antigens, seminal the protective effect of sperm exposure and the
TGF1 may inhibit the induction of Type 1 responses ‘dangerous’ partner are consistent with the ‘Immune
against the semi-allogenic conceptus that are thought maladaptation hypothesis’.
to be associated with poor placental development and Investigating the immunogenetic pathways
fetal. Decidual macrophages, present in an immune- potentially involved in preeclampsia, the human
suppressive phenotype from the moment of implanta- leukocyte antigen (HLA) system has been implicated.
tion, may inhibit NK cell lytic activity through their Many of the reported findings have been inconsistent
release of molecules such as TGF, Interleukin-10 (IL- or contradictory(28). HLA-G, a major histocommpati-
10) and prostaglandin-E2 (PGE2). Under the influence bility tissue specific antigen expressed in extravillous
of the local cytokine environment, antigen-presenting- trophoblast (EVT) cells (fetal derived), may protect
cells such as macrophages and dendritic cells may trophoblasts from maternal-fetal immune intolerance
take up, process and present ejaculate antigens and allow these cells to invade the uterus to ensure
(sperm, somatic cells, and soluble antigens) to T cells proper placentation(29). Extravillous trophoblast cells
in the draining lymph nodes loss(22,23). In mice, uptake express an unusual combination of HLA class I
of sperm mRNA encoding for paternal HLA by molecules: HLA-C, HLA-E and HLA-G. The maternal
decidual antigen presenting cells has been shown to decidua is infiltrated by a population of natural
occur, with subsequent translation of sperm mRNA killer (NK) cells with a distinctive phenotype. NK cells
encoding paternal MHC class I within these maternal typically function by cell killing or by cytokine
antigen presenting cells. These antigen-presenting production, which is enhanced by cytokines such as
cells traffic from the uterus to the draining lymph IFN a, IFN b, IL-2, IL-12 and IL-15(30). These cells are
nodes during the post-coital inflammatory response. particularly numerous in the decidua basalis at the
It is unknown whether or not this fascinating mecha- implantation site where they come into close contact
nism is operative in humankind. HLA-G is certainly with invading EVT cells. These NK cells express a
not involved here, since human sperm cells do not variety of receptors (CD94/NKG2, KIR and ILT) which
have mRNA for HLA-G(24,25). Since HLA-G is mono- are known to recognize HLA class I molecules. Inter-
morphic it would also be a very unlikely candidate to action between these NK cells and EVT cells might
represent the paternal HLA specificity. provide the controlling influence for implantation(31).
In summary, the primipaternity hypothesis Maternal uteroplacental blood flow increases
continues to stand strong. However, there might be during pregnancy. Altered uteroplacental blood flow
an additional effect associated with prolonged birth is a core predictor of abnormal pregnancy. Normally,
intervals. Structural changes of the spiral arteries the uteroplacental arteries are invaded by endo-
needed for the pregnancy, do not completely resolve vascular trophoblast and remodeled into dilated,
following parturition. The degree of anatomical inelastic tubes without maternal vasomotor control.
changes are related to the number of previous preg- Disturbed remodeling is associated with maintenance

J Med Assoc Thai Vol. 87 Suppl. 3 2004 S99

of high uteroplacental vascular resistance and intra- occur if the fetus is homozygous for the HLA-G null
uterine growth restriction (IUGR) and preeclampsia. allele, because the leader peptide will continue to be
Recent studies have shown that endovascular translated 30.
trophoblast invasion involves a side route of inter- As mentioned earlier, this activation of the
stitial invasion(32). Only humans have such extensive NK cells is probably of major importance in mediating
placental invasion, possibly because of the long intra- the major vascular adaptations.
uterine period that is required for the development of In summary, the major function of appro-
the fetal brain(30,33).The invasion of trophoblast cells priate interaction between trophoblast HLA and
into the decidua and the subsequent arterial transfor- uterine NK cells may not be simply to prevent lysis,
mation requires and results in close tissue contact but instead an active positive recognition resulting
between allogeneic cells. The mucosal lining of in major changes in decidual leukocyte cytokine
the uterus is transformed from endometrium in the production stimulating trophoblast growth, invasion
non-pregnant state to the decidua of pregnancy. and vascular remodeling. The term immunotrophism
Progesterone plays a major role in this transformation. has been used to describe this active process. The
A massive leukocyte infiltration is the major cellular great importance of these new insights is that we start
characteristic of this change(30). A true decidua is only to understand how pregnancy is based on an unique
formed in species that an invasive form of placentation. couple specific immune interaction not involving
The process begins in the luteal phase before T-cells. It also provides an explanation how previous
(potential) implantation. The uterine NK cells are sperm exposure could induce partner specific tolerance
probably derived from blood NK cells. During early (paternal HLA-C) translating in a more optimal
pregnancy the uterine NK cells accumulate as a placentation proving protection against preeclamp-
dense infiltrate around the invading cytotrophoblast sia. Future studies are required to confirm or refute
cells. From mid-gestation onwards they progressively this exciting hypothesis.
disappear. Therefore, their presence coincides
with the period of cytotrophoblast invasion, since The Mother
placentation in humans is complete by about 20 weeks The immunogenetic interaction between two
gestation(30). NK cells influence both trophoblast partners is important to determine the degree of
invasion and the actual maternal placental bed endovascular trophoblast invasion and as such the
vascular changes(30). The uterine NK cells produce a successfulness of placentation. There is however a
series of cytokines that are involved n angiogenesis clear and important maternal contribution that is
and vascular stability, including vascular endothelial probably of pivotal importance in determining the
growth factor (VEGF), placental growth factor (PlGF) magnitude of the maternal response syndrome. There
and angiopoietin 2 (ANG2). is a well described inherited maternal predisposition
The major ligands for the NK cell receptors to preeclampsia(34).
are the HLA-A, C and E. Both the invading tropho- The other side of the equation, besides the
blast cells and surrounding maternal cells express genes controlling the maternal inflammatory response,
HLA-E. This HLA-E may be of importance in preventing is the way the maternal cardiovascular system, and
lysis by NK cells by having preferential binding more specific the endothelium is coping with the
to NK cell inhibitory receptors(30). Interestingly, the exaggerated inflammatory response and the degree of
sequence of the monomer peptide derived from other increased shear stress. Factors having a significant
HLA proteins that is bound to HLA-E also influences negative impact on the coping capacity of the endo-
binding. Only the HLA-G leader sequence complexed thelium include: (1) age, (2) chronic hypertension, (3)
with HLA-E binds preferentially to activating NK thrombophilic disorders, (4) the homocysteine level,
cell receptors triggering an NK cell response(30). So, (5) obesity, (6) insulin resistance/hyperinsulinemia, (7)
uterine NK cells could respond differently to tropho- type I diabetes.
blast HLA-E complexed with HLA-G compared with
the surrounding HLA-E positive but HLA-G negative Genetic Conflict Hypothesis
maternal cells. In this way, HLA-G, which is expressed According to the genetic conflict theory(35,36),
only by extravillous trophoblast, could influence fetal genes will be selected to increase the transfer
the maternal immune response indirectly through of nutrients to the fetus, and maternal genes will be
presentation by HLA-E. This mechanism would still selected to limit transfers in excess of some maternal

S100 J Med Assoc Thai Vol. 87 Suppl. 3 2004

optimum. The phenomenon of genomic imprinting risk fetal strategy to increase nonplacental resistance
means that a similar conflict exists within fetal cells when a fetus’s uteroplacental blood supply is
between genes that are maternally derived and genes inadequate.
that are paternally derived. Endovascular trophobast The recent studies on soluble Flt provide
invasion has three consequences: (1) the fetus gains the molecular pathways predicted by Haig (37,38) .
direct access to its mother’s arterial blood. Therefore, Placental soluble fms-like tyrosine kinase 1 (sFlt1), an
a mother cannot reduce the nutrient content of blood antagonist of decidual (maternal) VEGF and placental
reaching the placenta without reducing the nutrient growth factor (PlGF), is upregulated in preeclampsia,
supply to her own tissues, (2) the volume of blood leading to increased systemic levels of sFlt1 that fall
reaching the placenta becomes largely independent after delivery 37. Increased circulating sFlt1 in patients
of control by the local maternal vasculature, and (3) with preeclampsia is associated with decreased
the placenta is able to release hormones and other circulating levels of free VEGF and PlGF, resulting in
substances directly into the maternal circulation. The endothelial dysfunction in vitro that can be rescued
conflict hypothesis predicts that placental factors by exogenous VEGF and PlGF. In a rat model VEGF
(fetal genes) will act to increase maternal blood and PlGF cause microvascular relaxation of rat renal
pressure, whereas maternal factors will act to reduce arterioles in vitro that is blocked by sFlt1, while
blood pressure. The conflict hypothesis suggests administration of sFlt1 to pregnant rats induces hyper-
that the mother reduces vascular resistance early in tension, proteinuria, and glomerular endotheliosis, the
pregnancy to ration fetal nutrients and that the sub- classic lesion of preeclampsia. Excess circulating sFlt1
sequent physiologic increase in vascular resistance may contribute to the pathogenesis of preeclampsia,
represents the changing “balance of power” as the and may represent an important fetal ‘rescue’ strategy
fetus grows larger. A corollary is that placental factors by increasing maternal blood pressure and as such
contribute to the increase in maternal cardiac output. uteroplacental perfusion pressure.
Placental factors have an opportunity to preferentially The conclusions derived from the studies
increase nonplacental resistance because the utero- reviewed in this paper may have practical consequences
placental arteries are highly modified and unrespon- for practicing physicians, even if the exact etiology of
sive to vasoconstrictors. The intrinsic effects of a preeclampsia remains unre-sol-ved:
high maternal systemic blood pressure are ultimately 1. According to the primipaternity concept,
beneficial to the fetus. Thus, genetic conflict hypo- a multiparous women with a new partner should be
thesis predicts that fetal genes will enhance the flow approached as being a primigravid women.
of maternal blood through the intervillous space by 2. Artificial donor insemination, oocyte
increasing maternal blood pressure (perfusion donation and especially embryo donation are
pressure). Maternal blood pressures form a continuum associated with an increased risk of developing
so that the dividing line between normotensive and pregnancy-induced hypertensive disor-ders.
hypertensive pregnancies is arbitrary. The conflict 3. A more or less prolonged period of sperm
hypothesis predicts that a mother’s position on this exposure provides a partial protection against preg-
continuum is determined by the balance between fetal nancy-induced hypertensive disorders. Nowadays all
factors increasing blood pressure and maternal women with changing partners are strongly advised to
factors decreasing blood pressure. This mechanism use condoms in order to prevent sexually transmitted
may be operative in gestational hypertension, where diseases. However, a certain period of sperm exposure
fetal prognosis is known to be good. In contrast, in within a stable relation, when pregnancy is aimed
preeclampsia the hypertension results from vasocon- for, is associated with a partial protection against
striction rather than increased cardiac output. An preeclampsia.
hypoxic placenta may release cytotoxic factors that
damage maternal endothelial cells, and thus cause References
vasoconstriction and an increase in maternal blood 1. Roberts JM, Redman CWG. Pre-eclampsia: More than
pressure. In this situation, small increments in pregnancy-induced hypertension. Lancet 1993; 341:
the birthweight of semistarved fetuses may often 1447-51.
have caused major increases in subsequent survival 2. Redman CWG, Sacks GP, Sargent IL. Preeclampsia:
despite substantial costs to the mother. Thus, endo- An excessive maternal inflammatory response to
thelial cell dysfunction may have evolved as a high pregnancy. Am J Obstet Gynecol 1999; 180: 499-506.

J Med Assoc Thai Vol. 87 Suppl. 3 2004 S101

 3. Dekker GA, Sibai BM. Etiology and Pathogenesis of 19. Lie RT, Rasmussen S, Brunborg H, Gjessing HK,
Preeclampsia: Current Concepts. Am J Obstet Gynecol Lie-Nielsen E, Irgens LM. Fetal and maternal
1998; 179: 1359-75. contributions to risk of pre-eclampsia: a population
4. Lorentzen B, Endresen MJ, Clausen T, Henriksen T. based study. Br Med J 1998; 316: 1343-7.
Fasting serum free fatty acids and triglycerides 20. Marti JJ, Herrmann U. Immunogestosis: a new
are increased before 20 weeks of gestation in women etiologic concept of “essential” EPH gestosis, with
who later develop preeclampsia. Hypertension in special consideration of the primigravid patient. Am J
Pregnancy 1994; 13: 103-9. Obstet Gynecol 1977;128:489-93.
5. Arbogast BW, Leeper SC, Merrick RD, Olive KE, 21. Klonoff-Cohen HS, Savitz DA, Cefalo RC, McCann
Taylor RN. Which plasma factors bring about MF. An epidemiologic study of contraception and
disturbance of endothelial function in pre-eclampsia? preeclampsia. JAMA 1989: 262: 3143-7.
Lancet 1994; 343: 340-1. 22. Tremellen KP, Seamark RF, Robertson SA, Seminal
6. Bosio PM, McKenna PJ, Conroy R, O’Herlihy. transforming growth factor 1 stimulates granulocyte-
Maternal central hemodynamics in hypertensive macrophage colony-stimulating factor production and
disorders of pregnancy. Obstet Gynecol 1999; 94: inflammatory cell recruitment in the murine uterus.
978-84. Biol Reprod 1998; 58: 1217-25.
7. Dekker GA. Risk factors for preeclampsia. Clin Obstet 23. Robertson SA. Transforming growth factor beta - a
Gynecol 1999; 42: 422-35. mediator of immune deviation in seminal plasma. J
8. Strickland DM, Guzick DS, Cox K, Gant NF, Rosenfeld Reprod Immunol. 2002; 57: 109-28.
CR. The relationship between abortion in the first 24. Watson JG, Carroll J, Chaykin S. Reproduction in mice:
pregnancy and development of pregnancy-induced the fate of spermatozoa not involved in fertilization.
hypertension in the subse-quent pregnancy. Am J Gamete Res 1983; 7: 75-84.
Obstet Gynecol 1986; 154: 1468. 25. Hiby SE, King A, Sharkey A, Loke YW. Molecular
9. Robillard PY, Hulsey TC, Alexander GR, Keenan A, studies of trophoblast HLA-G: polymorphisms,
de Caunes F, Papiernik E. Paternity patterns and risk isoforms, imprinting and expression in preimplanta-
of preeclampsia in the last preg-nancy in multiparae. tion embryo. Tissue Antigens 1999; 53: 1-13.
J Reprod Immunol 1993; 24: 112. 26. Khong TY, Adema ED, Erwich JJ. On an anatomical
10. Robillard PY, Hulsey TC, Perianin J, Janky E, Miri basis for the increase in birth weight in second and
EH, Papiernik E. Association of pregnancy-induced subsequent born children. Placenta. 2003; 24: 348-53.
hypertension with duration of sexual cohabitation 27. Robillard PY, Dekker GA, Hulsey TC. Evolutionary
before conception. Lancet 1994; 344: 973-5. adaptations to preeclampsia/eclampsia in humans:
11. Tubbergen P, Lachmeijer AM, Althuisius SM, Vlak low fecundability rate, loss of oestrus, prohibitions of
ME, van Geijn HP, Dekker GA. Change in paternity: incest and systematic polyandry. Am J Reprod
a risk factor for preeclampsia in multiparous women? Immunol 2002; 47: 104-11.
J Reprod Immunol 1999; 45: 81-8. 28. Kilpatrick DC. Influence of human leukocyte antigen
12. Dekker GA, Robillard PY, Hulsey TC. Immune and tumour necrosis factor genes on the development
maladaptation in the etiology of preeclampsia: a of pre-eclampsia. Hum Reprod Update 1999; 5: 94-102.
review of corroborative epidemiologic studies. Obstet 29. Goldman-Wohl DS, Ariel I, Greenfield C, Hochner-
Gynecol Surv 1998; 53: 377-82. Celnikier D, Cross J, Fisher S, et al. Lack of human
13. Robillard PY, Hulsey TC. Association of pregnancy- leukocyte antigen-G expression in extravillous
inducedhypertension, pre-eclampsia, and eclampsia trophoblasts is associated with preeclampsia. Mol
with duration of sexual cohabitation before conception. Hum Reprod 2000; 6: 88-95.
Lancet 1996; 347: 619. 30. Moffett-King A. Natural Killer Cells and Pregnancy
14. Wang JX, Knottnerus AM, Schuit G, Norman RJ, Chan Nat Rev Immunol 2002; 2: 656-63.
A, Dekker GA. Surgically obtained sperm, and risk of 31. Loke YW, King A. Immunology of implantation.
gestational hypertension and pre-eclampsia. Lancet Baillieres Best Pract Res Clin Obstet Gynaecol 2000;
2002; 359: 673-4. 14: 827-37.
15. Dekker GA. Oral tolerization to paternal antigens and 32. Kaufmann P, Black S, Huppertz B. Endovascular
preeclampsia. Am J Obstet Gynecol 1996; 174: 516. trophoblast invasion: implications for the patho-
16. Skjaerven R, Wilcox AJ, Lie RT. The interval between genesis of intrauterine growth retardation and
pregnancies and the risk of preeclampsia. N Engl J preeclampsia. Biol Reprod 2003; 69: 1-7.
Med 2002; 346: 33-8. 33. Robillard PY, Hulsey TC, Dekker GA, Chaouat G.
17. Macintyre S, Sooman A. Non-paterniy and prenatal Preeclampsia and human reproduction. An essay of
genetic screening. Lancet 1991; 338: 869-71. a long term reflection. J Reprod Immunol 2003; 59:
18. Li DK, Wi S. Changing paternity and the risk of 93-100.
preeclampsia/eclampsia in the subsequent pregnancy. 34. Lachmeijer A, Dekker G, Pals G, Aarnoudse J, ten
Am J Epidemiol 2000; 151: 57-62. Kate L, Arngrimsson R. Searching for preeclampsia

S102 J Med Assoc Thai Vol. 87 Suppl. 3 2004

 genes: the current position. Eur J Obstet Gynecol 37. Maynard SE, Min JY, Merchan J, Lim KH, Li J,
Reprod Biol 2002; 105: 94-113. Mondal S, et al. Excess placental soluble fms-like
35. Haig D. Genetic conflicts in human pregnancy.
tyrosine kinase 1 (sFlt1) maycontribute to endothelial
Quarterly Review of Biology 1993; 68: 495-532.
36. Haig D. Genomic imprinting and the theory of parent- dysfunction, hypertension, and proteinuria in
offspring conflict. Semin Devel Biol 1992; 3: 153-60. preeclampsia. Clin Invest 2003; 111: 649-58.

J Med Assoc Thai Vol. 87 Suppl. 3 2004 S103

View publication stats