Expert Reviews Obstetrics www.AJOG.

org

Pregnancy, parturition and preeclampsia in women

of African ancestry
Annettee Nakimuli, MBChB, MMed (Obs&Gyn); Olympe Chazara, PhD;
Josaphat Byamugisha, MBChB, MMed (Obs&Gyn), PhD; Alison M. Elliott, MA, MBBS, MD, DTM&H, FRCP;
Pontiano Kaleebu, MD, PhD; Florence Mirembe, MBChB, MMed (Obs&Gyn), PhD;
Ashley Moffett, MA, MB, BChir, MD, MRCP, MRCPath

States are 4, 12, and 21, respectively,

Maternal and associated neonatal mortality rates in sub-Saharan Africa remain unac- whereas those of Chad, Nigeria, and
ceptably high. In Mulago Hospital (Kampala, Uganda), 2 major causes of maternal death Congo are 1100, 630, and 540 per 100,000
are preeclampsia and obstructed labor and their complications, conditions occurring at live births, respectively. In SSA, the major
the extremes of the birthweight spectrum, a situation encapsulated as the obstetric direct causes of maternal mortality are
dilemma. We have questioned whether the prevalence of these disorders occurs more hemorrhage, preeclampsia/eclampsia,
frequently in indigenous African women and those with African ancestry elsewhere in the obstructed labor, and sepsis.4 Infections,
world by reviewing available literature. We conclude that these women are at greater risk preterm birth, birth asphyxia, stillbirths,
of preeclampsia than other racial groups. At least part of this susceptibility seems in- and small-for-gestational-age infants
dependent of socioeconomic status and likely is due to biological or genetic factors. are the leading causes of perinatal
Evidence for a genetic contribution to preeclampsia is discussed. We go on to propose mortality.2,5
that the obstetric dilemma in humans is responsible for this situation and discuss how These observations are representative
parturition and birthweight are subject to stabilizing selection. Other data we present also of our own institution, Mulago Hospital
suggest that there are particularly strong evolutionary selective pressures operating in Kampala (Uganda) in which the MMR
during pregnancy and delivery in Africans. There is much greater genetic diversity and has remained high at 438 per 100,000
less linkage disequilibrium in Africa, and the genes responsible for regulating birthweight live births, even though there has been
and placentation may therefore be easier to define than in non-African cohorts. Inclusion an increase in skilled birth attendance
of African women into research on preeclampsia is an essential component in tackling (58%) and very good attendance rate
this major disparity of maternal health. (95%) at antenatal clinics.6
Key words: evolutionary selective pressure, great obstetric syndromes, length of Mulago Hospital is the busiest ma-
gestation, obstetric dilemma ternity hospital in SSA, serving as a ter-
tiary referral center for Uganda. Details
of deliveries and maternal deaths are
shown in Table 1. Even with the lack

M ore than 90% of maternal deaths

worldwide occur in sub-Saharan
Africa (SSA) and south Asia. These high
partners, and others.1-3 The majority of
these deaths are related to pregnancy
complications that are inadequately
of good medical records that is charac-
teristic of much of SSA, our experience
in Mulago Hospital is that causes of
maternal and associated neonatal mor- managed because of a lack of access to maternal deaths are similar to the rest
tality rates persist despite considerable emergency health care. The maternal of SSA, with hemorrhage, preeclampsia/
efforts from the World Health Organi- mortality ratios (MMRs) of Sweden, eclampsia, and sepsis occurring very
zation, governments, development the United Kingdom, and the United commonly.4,7,8 The large number of

From the Department of Obstetrics and Gynaecology, Makerere University and Mulago Hospital, Kampala, Uganda (Dr Nakimuli and Drs Byamugisha and
Mirembe); Department of Pathology and Centre for Trophoblast Research, University of Cambridge, Cambridge, United Kingdom (Drs Chazara and
Moffett); Medical Research Council/Uganda Virus Research Institute Uganda Research Unit on AIDS, Entebbe, Uganda (Drs Elliott and Kaleebu); and
London School of Hygiene and Tropical Medicine, London, United Kingdom (Dr Elliott).
Received Sept. 11, 2013; revised Oct. 22, 2013; accepted Oct. 28, 2013.
This study was supported by the Wellcome Trust (grants 090108/Z/09/Z and 085992/Z/08/Z and a vacation scholarship in 2011); the British Heart
Foundation (grant PG/ 09/077/27964); the Centre for Trophoblast Research at the University of Cambridge; and a Wellcome Trust Uganda PhD
Fellowship in Infection and Immunity held by A.N., supported by a Wellcome Trust Strategic Award (grant 084344).
The authors report no conflict of interest.
Reprints: Ashley Moffett, MA, MB, BChir, MD, MRCP, MRCPath, Department of Pathology and Centre for Trophoblast Research, University of
Cambridge, Tennis Court Rd., Cambridge CB2 1QP, United Kingdom. am485@cam.ac.uk.
0002-9378/$36.00
ª 2014 Mosby, Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ajog.2013.10.879

510 American Journal of Obstetrics & Gynecology JUNE 2014

www.AJOG.org Obstetrics Expert Reviews
women seen with preeclampsia, partic-
ularly recurrent, severe, and early-onset TABLE 1
preeclampsia and eclampsia, is of par- Data for the maternity unit in Mulago Hospital, Uganda
ticular concern to us because these Variable 2009 2010 2011
conditions have a high mortality and Live births 30,247 31,585 32,633
morbidity, are impossible to predict,
Stillbirths 1260 1303 1230
and their pathogenesis is still somewhat
mysterious. Cesarean sections 6849 6702 6800
Here we review data relating to pre- Ruptured uterus 125 119 206
eclampsia in indigenous Africans and in Maternal deaths 187 152 188
women of African ancestry elsewhere
Attendance at antenatal clinic 78,157 76,673 69,129
in the world. We discuss the idea that in
Nakimuli. Pregnancy, parturition and preeclampsia in Africa. Am J Obstet Gynecol 2014.
these women, apart from the obvious,
cultural and socioeconomic factors and
different priorities in health care, there
are additional biological reasons why the
preeclampsia syndromes are such a prevalence of preeclampsia in SSA is Preeclampsia among African
prominent feature of African obstetrics. unknown because detailed clinical re- Americans
Our findings also lead us to question cords of all births are lacking. Dis- Although African Americans are obvi-
whether there are other characteristics of tinguishing between true preeclampsia ously not directly comparable with
pregnancy and parturition that differ in and pregnancy-induced hypertension is indigenous Africans because of consid-
African women. also difficult because proteinuria may erable genetic admixture (7-23%),18,19
not be adequately measured. A further the large number of reports and the
Preeclampsia and the great obstetric problem is a lack of information on consistency of the findings are informa-
syndromes preexisting hypertension because pre- tive (Table 2). For decades it has been
An important determinant of preeclamp- sentation to the clinic is often late. clear that there are disparities in obstet-
sia is failure of placentation, particularly Given this dearth of accurate records ric outcomes including preeclampsia
the physiological transformation of spi- of pregnancy outcomes in Uganda and between AA and other groups; indeed,
ral arteries, which leads to a stressed, SSA generally, to review the incidence black ethnicity is cited as a risk factor
underperfused placenta.9,10 Preeclamp- of preeclampsia in women of African for preeclampsia in reviews.20,21 Of 4
sia is one of a spectrum of pregnancy ancestry, we have reviewed reports re- million births recorded in the National
disorders that may result from this un- lating to preeclampsia in African Amer- Vital Statistics Report, pregnancy-
derlying pathogenesis, including fetal icans (AA) and immigrants from Africa associated hypertension was more com-
growth restriction (FGR), stillbirth, to other high-income countries as mon in AA (5.0%) and least frequent in
abruptio placentae, and some cases of compared with other ethnic groups. Hispanics (2.9%).22
preterm labor with intact membranes Studies were identified through a search A study of more than 2 million preg-
and prelabor rupture of membranes.11,12 of the PubMed database for relevant nancies in New York using data from
Because of the overlap in these condi- peer-reviewed articles published in En- hospital discharge records found that
tions, it is useful to think of them to- glish using the search terms, pre- the rates of preeclampsia were substan-
gether as the great obstetric syndromes eclampsia or eclampsia or hypertensive tially higher among AA compared with
(GOS) (Appendix).13-15 disease in pregnancy or gestational hy- European Americans. This was even
All these conditions are seen very pertension or severe maternal morbidity more obvious when confounders such as
frequently in Mulago Hospital. However, and ethnicity or race (Tables 2 and 3). diabetes and maternal age were taken
FGR cannot be reliably diagnosed In this review, we designate women into account. Furthermore, the differ-
without accurate knowledge of gesta- of African ancestry as those women ence persisted after stratification for
tional age, and low birthweight may descended from inhabitants of SSA. socioeconomic status based on area of
result from a variety of causes. Similarly, There are obvious caveats when reviewing residence.23
stillbirth is a heterogeneous condition data from women of African descent who Two other large studies in the United
that can result from congenital infection, have migrated to new environments. States, each with more than 1 million
birth asphyxia, or birth trauma as well as Those who have the energy to migrate women also found that preeclampsia was
poor uteroplacental perfusion. may be healthier than those left behind. more common in AA compared with
Because preeclampsia is a recognized Furthermore, factors such as diet, life- European Americans.24,25 One of these
clinical entity characterized by new style, education, health care, climate, and studies took data from the National
onset of hypertension and proteinuria indigenous pathogens are different and Inpatient Sample in which information
after 20 weeks’ gestation, we have necessarily become an integral part of was also available on health insurance
focused on this disorder.16,17 The exact the immigrant’s new environment. and income level; when this was taken

JUNE 2014 American Journal of Obstetrics & Gynecology 511

Expert Reviews Obstetrics www.AJOG.org

TABLE 2
Preeclampsia or eclampsia studies among African Americans
Preeclampsia or
Cohort size (total/AA) eclampsia, OR (95% CI) Comments Reference
2,571,069/450,098 1.67 (1.64e1.71)a PE in women in New York state 23

24
1,030,350/161,780 1.59 (1.49e1.69) Adjusted for maternal characteristics and obstetric history
25
1,472,912/420,576 1.30 (1.28e1.33) Adjusted for maternal characteristics and obstetric history
299,499/n.a. 1.39 (1.26e1.54)a Severe PE in women without chronic hypertension 26

27
206,428/19,512 2.12 (1.85e2.42) Adjusted for maternal characteristics and obstetric history
29
330/124 2.25 (0.88e5.78) Eclampsia, adjusted for maternal characteristics and obstetric history
30
4702/740 1.40 (1.20e1.80) Adjusted for maternal characteristics and obstetric history
31
271/38 2.50 (0.97e6.40) Adjusted for maternal characteristics and obstetric history
32
4314/1998 1.23 (0.88e1.72) Adjusted for maternal characteristics and obstetric history
33
153/35 2.27 (1.26e5.92) Late postpartum PE, not adjusted
34
2394/592 1.53 (1.00e2.35) Adjusted for maternal characteristics, obstetric history, and biochemical factors
103,860/13,748 1.36 (1.27e1.45)a PE in women with singleton birth at first delivery 35

36
2,770,871/121,017 1.81 (1.51e2.17) Eclampsia, adjusted for maternal characteristics and obstetric history
28
127,544/12,639 1.41 (1.25e1.62) Adjusted for maternal characteristics, chronic hypertension excluded
a 38
16,300/6000 1.63 (1.58e1.69) Eclampsia in racial minorities, not adjusted, not significant
40
1355/374 3.20 (1.04e9.93) PE in women without chronic hypertension
67
500/68 2.29 (1.16e4.53) Recurrent PE, adjusted for maternal characteristics and obstetric history
44
10,755/5555 1.30 (1.07e1.58) Adjusted for maternal characteristics
69
2947/156 1.62 (0.00e3.20) No effect when analyzed by recruitment center
Maternal characteristics generally include maternal age, body mass index, and smoking. Obstetric history generally includes parity, chronic hypertension, and diabetes.
CI, confidence interval; n.a., not available; OR, odds ratio; PE, preeclampsia.
a
OR was calculated from the data.
Nakimuli. Pregnancy, parturition and preeclampsia in Africa. Am J Obstet Gynecol 2014.

into account, the findings remained the for both eclampsia and preeclampsia revealed a higher incidence of pre-
same.24 The other study used data from in nulliparous and parous women have eclampsia and eclampsia among AA
women who were all Medicaid enrollees also shown AA are at higher risk women compared with their European
in 14 southern states.25 AA women were (Table 2).28-36 counterparts, irrespective of whether
also most likely to have other poor Confounding factors such as obesity, there was preexisting hypertension.41
maternal outcomes like preterm labor, preexisting chronic hypertension, and Investigation of the GOS other than
abruptio placenta, and stillbirth. diabetes are difficult to control for and preeclampsia is more difficult because
Another large study from the National are likely to contribute to the increased of the problems in accurate diagnosis
Hospital Discharge Survey found AA risk of preeclampsia among AA, partic- described above. Nonetheless, a consis-
women had a higher incidence of all ularly in the case of chronic hyperten- tent message is that ethnic disparities
hypertensive disorders in pregnancy and sion.37-39 That preeclampsia may not be exist for all the GOS (spontaneous pre-
a greater risk of severe complications of wholly explained by higher rates of term labor, FGR, stillbirth, and other
preeclampsia such as abruptio placenta chronic hypertension among AA women poor obstetric outcomes), and all have an
and stillbirth compared with European is suggested by a comparison between increased frequency among AA.26,42-45
Americans.26 Similar findings were made African and European Americans without In a study of more than 5 million
in a large Wisconsin study recruited from chronic hypertension; the prevalence of births comparing birth outcomes be-
hospital discharge data. AA women had hypertension in pregnancy was similar, tween US-born and foreign-born
the highest risk for all the different types but AA women still had an increased women, women of African ancestry
of preeclampsia when compared with diagnosis of preeclampsia.40 Similar had the highest rates of infant mor-
European American women.27 Several findings were made much earlier by the tality, low birthweight, and preterm
other studies looking at risk factors Collaborative Perinatal project, which births, whether US born or foreign

512 American Journal of Obstetrics & Gynecology JUNE 2014

www.AJOG.org Obstetrics Expert Reviews

TABLE 3
Preeclampsia studies among recent African immigrants to other countries
Cohort size Preeclampsia,
(total/Africans) OR or RR (95% CI) African origin Comments Location Reference
49
118,849/15,218 3.34 (2.25e4.96) Caribbean Adjusted for maternal characteristics Canada
49
118,849/9130 3.14 (2.04e4.83) SSA Adjusted for maternal characteristics Canada
50
2413/317 2.40 (1.10e5.60) SSA, Surinam, Univariate analysis The Netherlands
Antilles
51
2506/29 2.70 (1.20e6.20) Antilles Eclampsia in cases of SAMM, adjusted for The Netherlands
maternal characteristics and obstetric history
51
2506/90 6.20 (3.60e10.6) SSA Eclampsia in cases of SAMM, adjusted for The Netherlands
maternal characteristics and obstetric history
52
6215/331 2.06 (1.04e4.09) Cape Verde Adjusted for maternal characteristics and The Netherlands
obstetric history
52
6215/264 1.87 (0.86e4.06) Antilles Adjusted for maternal characteristics The Netherlands
and obstetric history
53
1728/576 2.47 (1.02e6.00) Ethiopia Standardized care between the groups Israel
compared
54
76,158/11,395 2.60 (2.32e2.92) Caribbean Adjusted for maternal characteristics United Kingdom
and obstetric history
55
8366/1581 3.64 (1.84e7.21) n.a. Adjusted for maternal characteristics United Kingdom
and obstetric history
15,639/356 0.90 (0.53e1.51)a SSA No increased risk Sweden 56

58
165,001/986 n.a. African, Somalia No increased risk Finland
526/158 3.90 (1.70e8.94)a SSA Early-onset PE compared to late onset France 68

(<28 or <34 weeks)

Maternal characteristics generally include maternal age, body mass index, and socioeconomic status. Obstetric history generally includes parity, chronic hypertension, and diabetes.
CI, confidence interval; n.a., not available; OR, odds ratio; PE, preeclampsia; RR, relative risk; SAMM, severe acute maternal morbidity.
a
OR was calculated from the data.
Nakimuli. Pregnancy, parturition and preeclampsia in Africa. Am J Obstet Gynecol 2014.

born.46 In addition, the risk of pre- outcomes also apply to the Hispanic because these births often take place
term birth, stillbirth, and low birth- population in the United States, yet in countries with good records and uni-
weight is increased not only in AA several studies have noted that pre- versal health care systems (Table 3). For
women but also with AA fathers.47,48 eclampsia, low birthweight, and still- example, a large study of more than
Explanations for the disparities found birth are similar or even better than for 100,000 women who immigrated to
between women with African or Euro- white women, the Hispanic paradox.28,44 Ontario between 1985 and 2000 showed
pean ancestry have been poor socioeco- Using information from the Duke Uni- that the racial groups with the highest
nomic status with lower incomes and versity Birth Database, AA women had risk of severe preeclampsia were from
level of education, lack of medical in- higher rates of preeclampsia (10.2%) the Caribbean or SSA.49
surance, poor utilization of precon- than the European (8%) or Hispanic Similar findings were made in The
ception and antenatal services, stress, women (6.2%), even though the socio- Netherlands where the highest risk for
discrimination, and residential segre- economic status of Hispanic and AA eclampsia and preeclampsia was from
gation. Several reports have tried to women was similar.44 women from SSA.50,51 Cape Verdean and
determine the impact of these factors; Antillean women were also at higher risk
for example, women of African ancestry Preeclampsia among more recent of preeclampsia in a report from Rotter-
were at an increased risk of pre- African immigrants to other countries dam, The Netherlands.52 A large number
eclampsia in a second pregnancy, but Large numbers of Africans have migrated of Ethiopians have settled in Israel since
this was not associated with Medicaid to Europe and other high-income coun- the 1980s where prenatal and obstetric
enrollment.30 tries, mainly in the past 50 years. Ob- care is standardized with equal medical
Many of the socioeconomic factors stetric outcomes for these recent African insurance, and in this group severe pre-
that may contribute to poor obstetric immigrants are informative, particularly eclampsia was more likely to occur.53

JUNE 2014 American Journal of Obstetrics & Gynecology 513

Expert Reviews Obstetrics www.AJOG.org

Large groups of women of African Increased risk of other GOS, even if pregnancy, are also at increased risk.73-77
ancestry live in London where access preeclampsia does not occur, is also These findings of familial aggregation in
to National Health Service hospitals is clear from another large study in Swe- preeclampsia are also true for the other
freely available and home deliveries are den, and other reports support this GOS.78-81 Although environmental fac-
rare. In a survey of 80,000 pregnancies conclusion.35,63-66 tors, particularly influences acting in
that included women of European and Large studies of this kind are still not utero, are important, some of the risk is
Asian ancestry, 15% had African an- available for African women resident in likely to be genetic. Indeed, a study of
cestry, and this was the second strongest SSA, but our own experience in Kampala female twin pairs with known zygosity
risk factor for preeclampsia after chronic is that recurrent preeclampsia does occur estimated that the heritability of pre-
hypertension and also carried a higher frequently. In London, 23% of 500 eclampsia was approximately 54%.82
risk of other poor obstetric outcomes women with previous preeclampsia had Could there be particular susceptibility
such as FGR and stillbirth.54 Similarly, recurrent disease that required delivery genes associated with the higher fre-
African ancestry was a risk factor for before 37 weeks, and African compared quency of preeclampsia in women of
early-onset preeclampsia compared with with European ancestry was a significant African ancestry? A case-control study of
all other racial groups, and this remained predictor.67 It also seems that when preeclampsia in Latinas, a group with
so, even after adjusting for age, body preeclampsia does occur in the second admixture from European, African, and
mass index, and other maternal charac- pregnancy in AA women, it is severe, native Americans, did show, using
teristics.55 These and other studies of early-onset disease with associated FGR ancestry informative markers, that Afri-
African immigrants also highlight the and preterm birth.45 A recent study from can ancestry was associated with
increased risk of GOS such as stillbirths France suggests that women of African preeclampsia.83
and FGR similar to AA.56-58 ancestry are more at risk for early-onset The role of the fetal (father’s) genes is
preeclampsia and more likely to have less obvious, but many reports indicate a
Severity and recurrence had a previous history of preeclampsia paternal contribution to the risk.84-87
of preeclampsia compared with other groups, including Intergenerational and familial aggrega-
The early onset and severity of pre- women from North Africa, despite the tion also point to genetic factors derived
eclampsia in women from Uganda is also even higher incidence of chronic hyper- from both maternal and fetal genes, with
a cause of concern, although the latter tension in the latter group.68 most risk coming from maternal genes
may reflect the late admittance to Mulago that may act in either the mother or her
Hospital. In a US national hospital Summary fetus.71,74,77,88 A drive to look for the
discharge survey, higher mortality from Our comprehensive review of the liter- susceptibility genes for preeclampsia has
preeclampsia and eclampsia was reported ature identified very few papers that run so far been disappointing. The studies
among women of African ancestry counter to our conclusion that women of generally have small numbers of subjects
compared with European Americans, but African ancestry are at increased risk of and have not been replicated.89
only one-third or less of the difference developing preeclampsia. First, 3 studies Genome-wide association screening is
could actually be attributed to the higher showed that these women were not at an unbiased approach to look for sus-
prevalence.59 Pregnancy-related deaths increased risk of preeclampsia, but they ceptibility genes in complex disorders and
from preeclampsia/eclampsia were 3 had low power to detect any effect.56,58,69 has been used in preeclamptic cohorts,
times higher in AA women compared Second, the apparent increased suscep- but, although various single-nucleotide
with Europeans.60 tibility to preeclampsia among AA has polymorphism candidates have been
In the UK Maternal Death Review been dismissed as a problem of incorrect identified, the lack of statistical power is
for the period 2006-2008, 22 deaths diagnosis.37 Third, race was discounted again a problem.90-93 Systematic meta-
occurred as a result of preeclampsia and as a significant risk factor for pre- analyses of these studies found 7 single-
eclampsia. Despite being a minority eclampsia in another study, but data nucleotide polymorphisms significantly
group, 6 of these deaths were Africans regarding AA women were combined associated near genes involved in pro-
and the authors noted: “Black African with that for other minority races so that cesses such as coagulation, the renin-
women seem particularly susceptible to the analysis could not provide a mean- angiotensin system, and inflammation.
aggressive forms of preeclampsia. To ingful comparison.38 This highlights an important issue:
establish if this is true, and what might searching for variants associated with
be the underlying genetic or other Genetics of preeclampsia preeclampsia only in the maternal ge-
pathophysiological mechanisms, further That there is a genetic component nome will reveal genes mainly associated
research is required.”61 to preeclampsia has long been sus- with the tertiary systemic syndrome and
After a woman has had preeclampsia pected.70-72 Daughters of women with not those maternal and/or fetal genes
in her first pregnancy, the risk of recur- preeclampsia have more than twice the involved in physiological transformation
rence is increased, with a relative risk of risk of developing the disease them- of the arteries or to the subsequent stress
15.0 cited in an authoritative Norwegian selves, and sisters of affected women, response of the placenta to the reduced
study of more than 2 million women.62 even if not born from a preeclamptic blood flow. The clear increased risk of

514 American Journal of Obstetrics & Gynecology JUNE 2014

www.AJOG.org Obstetrics Expert Reviews
cardiovascular disease in women who
have had preeclampsia again points to a FIGURE 1
separate set of susceptibility genes that Maternal KIR/fetal HLA-C interactions at the site of placentation
are acting systemically and not during
early placentation.94,95
We have taken a different approach and
focused on the primary defect of poor
placentation. This is based on the idea
that regulation of trophoblast behavior
during placentation is mediated by allo-
geneic recognition of trophoblast major
histocompatibility complex molecules by
maternal lymphocytes.96 The findings
that specialized immune cells, uterine
natural killer (NK) cells, accumulate at
the site of placentation, together with the
discovery of NK receptors, the killer-cell
immunoglobulin-like receptor family
(KIR) and their cognate HLA-C tropho-
blast ligands have demonstrated how the
mother can discern the presence of a
genetically different individual.97-99
KIR and HLA are the most poly-
morphic gene families in humans, and
we have shown that particular maternal
KIR in combination with fetal HLA-C
variants are associated with preeclampsia
and the other GOS.100-102 Women who
have 2 KIR A haplotypes (KIR AA geno-
type) are at risk when there is a HLA-C
allele belonging to the C2 group in
the fetus. Furthermore, the origin of the
fetal HLA-C2 is important; the most risk is
from a C2 allele inherited from the father.
We are now undertaking a similar
study at Mulago Hospital, and pre- In these 2 scenarios, the mother is HLA-C1 homozygous and the fetus has inherited an HLA-C2
liminary findings illustrate the same group allele from the father. If the mother has a KIR AA genotype that lacks activating KIR and has a
maternal KIR/fetal HLA-C combinations strong inhibitory KIR for HLA-C2 (KIR2DL1), poor placentation results. In contrast, if the mother has a
associated with preeclampsia in African KIR AB or BB genotype containing the activating KIR for HLA-C2 (KIR2DS1), uterine natural killer cells
women. Interestingly, the frequency of are triggered to produce increased amount of cytokines and chemokines (eg, granulocyte-
the fetal HLA-C2 variant that confers macrophage colonyestimulating factor) that enhance placentation.
risk is increased in Ugandans compared Nakimuli. Pregnancy, parturition and preeclampsia in Africa. Am J Obstet Gynecol 2014.
with Europeans and Asians.103 Further-
more, there is enormous variability of
KIR genes in Africans with far more ge- women who have a protective KIR B receives sufficient nourishment for nor-
notypes and more allelic variation at haplotype (in which the activating KIR mal development through remodeling of
individual KIR loci.104 for HLA-C2, KIR2DS1, is located) results the spiral arteries.
How these genetic findings translate in secretion of cytokines and chemokines
into the function of uterine NK cells is a that may facilitate trophoblast invasion The obstetric dilemma
challenge, given the ethical and logistical and vascular transformation.105 Thus, we Overall, the data we have brought
difficulties in experimenting with these propose that the uterine immune system together in this review suggest that pre-
cells. Functionally, we would predict that using highly variable maternal KIR/fetal eclampsia and other GOS occur more
the risky combination results in very HLA-C interactions subtly defines the commonly in women of African ancestry
strong inhibition of uterine NK cells boundary between mother and baby, compared with other ethnic groups, and
(Figure 1). Triggering of uterine NK cells limiting the highly invasive placenta while this is not wholly accounted for by con-
by HLA-C2 target cells in vitro from at the same time ensuring the fetus founding social, cultural, and medical

JUNE 2014 American Journal of Obstetrics & Gynecology 515

Expert Reviews Obstetrics www.AJOG.org

pelvic trauma, sepsis, and long-term

FIGURE 2 problems such as vesicovaginal fis-
Birthweight and neonatal mortality rates (n [ 13,730). tula.109 In Uganda, 2% of all women
have had an obstetric fistula.6 Therefore,
the higher cesarean section rates seen in
high-income countries in women of
African ancestry may reflect not just
delivery of women with preeclampsia
but also an increased frequency of
obstructed labor.24,54,58,110
A detailed audit from the Royal Col-
lege of Obstetricians and Gynaecologists
in the United Kingdom highlighted the
higher cesarean section rates in women
of African ancestry, even when con-
founders such as age, parity, birthweight,
and presentation were considered.110
Furthermore, shoulder dystocia has also
been reported to occur more commonly
in AA women.111
These findings may in part be
accounted for by measurements of the
bony pelvis revealing that there is even
less room for the fetal head in women of
African ancestry. Although pelvimetry
may not be a useful indicator in pre-
dicting cephalopelvic disproportion in
Adapted, with permission, from Cavalli-Sforza and Bodmer.108
Nakimuli. Pregnancy, parturition and preeclampsia in Africa. Am J Obstet Gynecol 2014.
individual patients,112 the measurements
made of the pelvis, notably the pelvic
inlet, outlet, length of sacrum, and pelvic
influences.50,54,57,68 It also seems that the Compared with other primates, the floor area, are all smaller in women of
risk of preeclampsia in African immi- passage of the large human fetal head African ancestry compared with those of
grants to Europe is increased irrespective through a bony pelvis is a tight fit, European ancestry.113-115
of their area of origin in Africa, apart requiring rotation of the head as it goes A possible consequence is that the
from North Africans.68 All these obser- through the birth canal as a consequence fetal head engages into the pelvis late,
vations point to a need to investigate of adaptation to bipedalism.106,107 The only when labor commences, whereas
possible biological/genetic reasons con- high maternal and neonatal mortality this occurs in the last month of gestation
tributing to the higher risk of pre- associated with extremes of birthweight, in European and Asian women.116
eclampsia in SSA. sometimes called the obstetric dilemma, Several reports also document the fact
We would anticipate that there would has been described as “perhaps the that normal term in African pregnancy
be strong selective pressure against a most clear-cut example of a human occurs at only approximately 38 weeks’
disorder that, without medical interven- character subject to stabilizing selection” gestation, which is 2 weeks earlier than
tion, is frequently fatal to mother and (Figure 2).108 in non-Africans, possibly facilitating
child and occurs in 5-10% of first preg- The optimal survival of babies weigh- birth before the baby becomes too
nancies. A failure of the physiological ing between 6 and 8 lb (2.5-3.5 kg) seems big.117-120 If births are occurring earlier,
transformation of uterine arteries is a to be a universal feature of human pop- it would certainly then be advantageous
common feature of all the GOS, and this ulations. If babies become too large, the for the fetus to mature more quickly,
results in a reduced placental supply risk of obstructed labor is increased. As and indeed, African babies frequently
of oxygen and nutrients, lower birth- in the rest of SSA, at Mulago Hospital we pass meconium with no sign of fetal
weights, and the risk of preterm labor not only have many disordered preg- distress.119,121 There is also accelerated
and superimposed preeclampsia. How- nancies arising from failure of placenta- lung maturity.122,123 Thus, the diffi-
ever, at the same time, we have to consider tion, but we also experience frequent culties imposed by the birth process and
that maternal and neonatal mortality is births with prolonged obstructed labor the passage of the head and shoulders
not only high under circumstances of because of cephalopelvic disproportion. through the bony pelvis seem to have
reduced fetal nutrition but also when Without cesarean section, this leads to driven biological changes in many as-
babies are too large for the pelvis. birth asphyxia, postpartum hemorrhage, pects of pregnancy.

516 American Journal of Obstetrics & Gynecology JUNE 2014

www.AJOG.org Obstetrics Expert Reviews
This clear trade-off between the size of out-of-Africa migrations have reduced maternal death: a systematic review. Lancet
the pelvis and head room will benefit the extent of genetic variation in the Eu- 2006;367:1066-74.
5. Lawn JE, Cousens S, Zupan J. Four million
babies who have intermediate birth- ropean populations that are the focus of neonatal deaths: when? where? why? Lancet
weights, and the 2 extremes with re- the great majority of studies of pregnancy 2005;365:891-900.
duced survival will be selected against. disorders. 6. Uganda Bureau of Statistics (UBOS) and ICF
Of interest will be a study of maternal Studying biological diversity will shed International Inc.Uganda Demographic and
KIR/fetal HLA-C variants in pregnancies light on pathological pregnancies in all Health Survey 2011, Kampala, Uganda. Kam-
pala (Uganda): Ganda Bureau of Statistics
with obstructed labor as the KIR A and B populations, and inclusion of African (UBOS) and ICF International Inc; 2012.
haplotypes and HLA-C1 and C2 groups women into research on preeclampsia 7. Kaye D, Mirembe F, Aziga F, Namulema B.
are under stabilizing selection and are is an essential component in tackling Maternal mortality and associated near-misses
found at different frequencies across the this major disparity of maternal health. It among emergency intrapartum obstetric re-
world’s populations.124 has been highlighted that “Africa is a ferrals in Mulago Hospital, Kampala, Uganda.
East Afr Med J 2003;80:144-9.
Therefore, when considering why pre- genetically special place” with greater ge- 8. Wandabwa JN, Doyle P, Longo-Mbenza B,
eclampsia persists in populations, it is netic diversity and lower levels of linkage et al. Human immunodeficiency virus and
important not only to consider the GOS disequilibrium.126 The genes contributing AIDS and other important predictors of
but also the morbidity and mortality at to defective placentation are therefore maternal mortality in Mulago Hospital Com-
the other end of the birth weight distri- likely to become obvious more quickly. plex Kampala Uganda. BMC Public Health
2011;11:565.
bution. It may be that the consequences Indeed, our initial study of KIR genes has 9. Redman CWG, Sargent IL. Immunological
of these contrasting selective pressures found a much greater number of different factors and placentation: implications for pre-
not only affect higher birthweight babies KIR haplotypes in women delivering at eclampsia. In: Lyall F, Belfort M, eds. Pre-
but also the tendency for enhanced Mulago Hospital but also highlighted the eclampsia. Cambridge (United Kingdom):
numbers of undernourished babies with extent of variability of these genes within Cambridge University Press; 2007.
10. Redman CW, Sargent IL. Latest advances
the concomitant maternal syndrome of the African continent.103 Denying the in understanding preeclampsia. Science 2005;
preeclampsia. That is to say that selection existence of genetic differences in Africans 308:1592-4.
in a population for reduced fetal size may and their interactions with nongenetic 11. Ananth CV, Vintzileos AM. Ischemic placental
lead to the persistence of factors predis- factors only delays the identification of the disease: epidemiology and risk factors. Eur J
posing to preeclampsia. causal genes or alleles that would allow us Obstet Gynecol Reprod Biol 2011;159:77-82.
12. Khong TY, De Wolf F, Robertson WB,
to move away from racial/ethnic catego- Brosens I. Inadequate maternal vascular
Future directions rization of individuals. Knowing the ge- response to placentation in pregnancies
There is an urgent need to document netic variants will allow a better complicated by pre-eclampsia and by small-for-
obstetric events better in SSA, and the understanding of the molecular pathways gestational age infants. Br J Obstet Gynaecol
lack of detailed electronic hospital re- and better health care for the women 1986;93:1049-59.
13. Brosens I, Pijnenborg R, Vercruysse L,
cords in the hospitals as well as the fail- carrying the risky genotypes, indepen- Romero R. The “Great Obstetrical Syndromes”
ure to record all births in the population dently from their ethnicity. - are associated with disorders of deep placenta-
is a major difficulty for any maternal tion. Am J Obstet Gynecol 2011;204:193-201.
health research program. For example, ACKNOWLEDGMENTS 14. Romero R, Kusanovic JP, Kim CJ. Placental
proper assessment of gestational age at bed disorders in the genesis of the great
We thank all the clinical and research staff of the obstetrical syndromes. In: Brosens I, Pijnenborg
delivery will be crucial in the accurate Mulago Hospital and the patients and their R, Romero R, eds. Placental bed disorders:
diagnosis of the GOS. The record systems families who made this work possible. We are basic science and its translation to obstetrics.
documenting obstetric and neonatal also grateful to Graham Burton, Hilary Critchley, Cambridge, NY: Cambridge University Press;
David Dunne, and Christopher Redman for their
problems are still inadequate throughout 2010:271-89.
helpful comments on the draft manuscript. 15. Smith GC. First-trimester determination of
SSA, and this is clearly a major priority.
complications of late pregnancy. JAMA
Introduction of record systems in 2010;303:561-2.
Zambia had an immediate impact on 16. North R. Classification and diagnosis of pre-
health care such as penicillin adminis- REFERENCES eclampsia. In: Lyall F, Belfort M, eds.
tration for syphilis as well as highlighting 1. World Health Organization. Trends in Pre-eclampsia. Cambridge (United Kingdom):
the shorter gestational age and describing maternal mortality: 1990 to 2008. Geneva Cambridge University Press; 2007.
(Switzerland): World Health Organization; 2010. 17. Cunningham FG, Leveno KJ, Bloom SL,
the normal birthweight distribution.125
2. World Health Organization. The World Health Hauth JC, Gilstrap L, Wenstrom KD. Hyperten-
The majority of studies on pre- Report 2005. Make every mother and child sive disorders in pregnancy. Williams obstetrics.
eclampsia and the other GOS are from count. Geneva (Switzerland): World Health Or- New York: McGraw Hill; 2005.
Europe and the Americas, but if these ganization; 2005. 18. Reed TE. Caucasian genes in American
conditions are indeed occurring at greater 3. World Health Organization. Reduction of Negroes. Science 1969;165:762-8.
maternal mortality: a joint WHO/UNFPA/ 19. Chakraborty R, Kamboh MI, Ferrell RE.
frequency in women of African ancestry, “Unique” alleles in admixed population: a strat-
UNICEF/World Bank statement. Geneva
it makes scientific and economic sense to (Switzerland): World Health Organization; 1999. egy for determining “hereditary” population dif-
study them in the setting in which they 4. Khan KS, Wojdyla D, Say L, Gulmezoglu AM, ferences of disease frequencies. Ethn Dis
have a major impact. Furthermore, the Van Look PF. WHO analysis of causes of 1991;1:245-56.

JUNE 2014 American Journal of Obstetrics & Gynecology 517

Expert Reviews Obstetrics www.AJOG.org

20. Dekker GA. Risk factors for preeclampsia. 37. Chesley LC. History and epidemiology of 52. Bouthoorn SH, Gaillard R, Steegers EA,
Clin Obstet Gynecol 1999;42:422-35. preeclampsia-eclampsia. Clin Obstet Gynecol et al. Ethnic differences in blood pressure and
21. Roberts JM, Cooper DW. Pathogenesis and 1984;27:801-20. hypertensive complications during pregnancy:
genetics of pre-eclampsia. Lancet 2001;357: 38. Saftlas AF, Olson DR, Franks AL, Atrash HK, the Generation R study. Hypertension 2012;60:
53-6. Pokras R. Epidemiology of preeclampsia and 198-205.
22. Martin JA, Hamilton BE, Ventura SJ, et al. eclampsia in the United States, 1979-1986. Am 53. Salim R, Mfra A, Garmi G, Shalev E. Com-
Births: final data for 2009. Natl Vital Stat Rep J Obstet Gynecol 1990;163:460-5. parison of intrapartum outcome among immi-
2011;60:1-70. 39. Samadi AR, Mayberry RM, Zaidi AA, grant women from Ethiopia and the general
23. Tanaka M, Jaamaa G, Kaiser M, et al. Racial Pleasant JC, McGhee N Jr, Rice RJ. Maternal obstetric population in Israel. Int J Gynaecol
disparity in hypertensive disorders of pregnancy hypertension and associated pregnancy com- Obstet 2012;118:161-5.
in New York State: a 10-year longitudinal plications among African-American and other 54. Khalil A, Rezende J, Akolekar R,
population-based study. Am J Public Health women in the United States. Obstet Gynecol Syngelaki A, Nicolaides KH. Maternal racial
2007;97:163-70. 1996;87:557-63. origin and adverse pregnancy outcome: a
24. Shen JJ, Tymkow C, MacMullen N. Dis- 40. Bryant AS, Seely EW, Cohen A, cohort study. Ultrasound Obstet Gynecol
parities in maternal outcomes among four ethnic Lieberman E. Patterns of pregnancy-related 2013;41:278-85.
populations. Ethn Dis 2005;15:492-7. hypertension in black and white women. 55. Poon LC, Kametas NA, Chelemen T,
25. Zhang S, Cardarelli K, Shim R, Ye J, Hypertens Pregnancy 2005;24:281-90. Leal A, Nicolaides KH. Maternal risk factors for
Booker KL, Rust G. Racial disparities in eco- 41. Niswander KR, Gordon M. The women and hypertensive disorders in pregnancy: a multi-
nomic and clinical outcomes of pregnancy their pregnancies: the collaborative perinatal variate approach. J Hum Hypertens 2010;24:
among Medicaid recipients. Matern Child Health study of the National Institute of Neurological 104-10.
J 2013;17:1518-25. Diseases and Stroke. Philadelphia, PA: W. B. 56. Essen B, Hanson BS, Ostergren PO,
26. Zhang J, Meikle S, Trumble A. Severe Saunders Co; 1972. Lindquist PG, Gudmundsson S. Increased
maternal morbidity associated with hypertensive 42. MacDorman MF. Race and ethnic dispar- perinatal mortality among sub-Saharan immi-
disorders in pregnancy in the United States. ities in fetal mortality, preterm birth, and infant grants in a city-population in Sweden. Acta
Hypertens Pregnancy 2003;22:203-12. mortality in the United States: an overview. Obstet Gynecol Scand 2000;79:737-43.
27. Cabacungan ET, Ngui EM, McGinley EL. Semin Perinatol 2011;35:200-8. 57. Jenum AK, Sommer C, Sletner L, Morkrid K,
Racial/ethnic disparities in maternal morbidities: 43. Bryant AS, Worjoloh A, Caughey AB, Baerug A, Mosdol A. Adiposity and hyper-
a statewide study of labor and delivery hospi- Washington AE. Racial/ethnic disparities in ob- glycaemia in pregnancy and related health out-
talizations in Wisconsin. Matern Child Health J stetric outcomes and care: prevalence and comes in European ethnic minorities of Asian
2012;16:1455-67. determinants. Am J Obstet Gynecol 2010;202: and African origin: a review. Food Nutr Res
28. Caughey AB, Stotland NE, Washington AE, 335-43. 2013:57.
Escobar GJ. Maternal ethnicity, paternal 44. Brown HL, Chireau MV, Jallah Y, Howard D. 58. Malin M, Gissler M. Maternal care and birth
ethnicity, and parental ethnic discordance: pre- The “Hispanic paradox”: an investigation of outcomes among ethnic minority women in
dictors of preeclampsia. Obstet Gynecol racial disparity in pregnancy outcomes at a ter- Finland. BMC Public Health 2009;9:84.
2005;106:156-61. tiary care medical center. Am J Obstet Gynecol 59. Tucker MJ, Berg CJ, Callaghan WM, Hsia J.
29. Abi-Said D, Annegers JF, Combs- 2007;197:197.e1-7; discussion 97.e7-9. The black-white disparity in pregnancy-related
Cantrell D, Frankowski RF, Willmore LJ. Case- 45. Mbah AK, Alio AP, Marty PJ, Bruder K, mortality from 5 conditions: differences in prev-
control study of the risk factors for eclampsia. Wilson R, Salihu HM. Recurrent versus isolated alence and case-fatality rates. Am J Public
Am J Epidemiol 1995;142:437-41. pre-eclampsia and risk of feto-infant morbidity Health 2007;97:247-51.
30. Mostello D, Catlin TK, Roman L, outcomes: racial/ethnic disparity. Eur J Obstet 60. MacKay AP, Berg CJ, Atrash HK. Preg-
Holcomb WL Jr, Leet T. Preeclampsia in the Gynecol Reprod Biol 2011;156:23-8. nancy-related mortality from preeclampsia and
parous woman: who is at risk? Am J Obstet 46. Singh GK, Yu SM. Adverse pregnancy out- eclampsia. Obstet Gynecol 2001;97:533-8.
Gynecol 2002;187:425-9. comes: differences between US- and foreign- 61. Cantwell R, Clutton-Brock T, Cooper G,
31. Eskenazi B, Fenster L, Sidney S. born women in major US racial and ethnic et al. Saving mothers’ lives: reviewing maternal
A multivariate analysis of risk factors for pre- groups. Am J Public Health 1996;86:837-43. deaths to make motherhood safer: 2006-2008.
eclampsia. JAMA 1991;266:237-41. 47. Palomar L, DeFranco EA, Lee KA, The Eighth Report of the Confidential Enquiries
32. Sibai BM, Ewell M, Levine RJ, et al. Risk Allsworth JE, Muglia LJ. Paternal race is a risk Into Maternal Deaths in the United Kingdom.
factors associated with preeclampsia in healthy factor for preterm birth. Am J Obstet Gynecol BJOG 2011;118(Suppl 1):1-203.
nulliparous women. The Calcium for Pre- 2007;197:152.e1-7. 62. Klungsoyr K, Morken NH, Irgens L,
eclampsia Prevention (CPEP) Study Group. Am 48. Srinivasjois RM, Shah S, Shah PS. Biracial Vollset SE, Skjaerven R. Secular trends in the
J Obstet Gynecol 1997;177:1003-10. couples and adverse birth outcomes: a sys- epidemiology of pre-eclampsia throughout 40
33. Larsen WI, Strong JE, Farley JH. Risk fac- tematic review and meta-analyses. Acta Obstet years in Norway: prevalence, risk factors and
tors for late postpartum preeclampsia. J Reprod Gynecol Scand 2012;91:1134-46. perinatal survival. Paediatr Perinat Epidemiol
Med 2012;57:35-8. 49. Urquia ML, Ying I, Glazier RH, Berger H, De 2012;26:190-8.
34. Myatt L, Clifton RG, Roberts JM, et al. First- Souza LR, Ray JG. Serious preeclampsia 63. Hjartardottir S, Leifsson BG, Geirsson RT,
trimester prediction of preeclampsia in nullipa- among different immigrant groups. J Obstet Steinthorsdottir V. Recurrence of hypertensive
rous women at low risk. Obstet Gynecol Gynaecol Can 2012;34:348-52. disorder in second pregnancy. Am J Obstet
2012;119:1234-42. 50. Knuist M, Bonsel GJ, Zondervan HA, Gynecol 2006;194:916-20.
35. Mostello D, Kallogjeri D, Tungsiripat R, Treffers PE. Risk factors for preeclampsia in 64. van Rijn BB, Hoeks LB, Bots ML, Franx A,
Leet T. Recurrence of preeclampsia: effects of nulliparous women in distinct ethnic groups: a Bruinse HW. Outcomes of subsequent pregnancy
gestational age at delivery of the first pregnancy, prospective cohort study. Obstet Gynecol after first pregnancy with early-onset preeclamp-
body mass index, paternity, and interval between 1998;92:174-8. sia. Am J Obstet Gynecol 2006;195:723-8.
births. Am J Obstet Gynecol 2008;199:55.e1-7. 51. Zwart JJ, Jonkers MD, Richters A, et al. 65. Wikstrom AK, Stephansson O, Cnattingius S.
36. Fong A, Chau CT, Pan D, Ogunyemi DA. Ethnic disparity in severe acute maternal Previous preeclampsia and risks of adverse
Clinical morbidities, trends, and demographics morbidity: a nationwide cohort study in The outcomes in subsequent nonpreeclamptic
of eclampsia: a population-based study. Am J Netherlands. Eur J Public Health 2011;21: pregnancies. Am J Obstet Gynecol 2011;204:
Obstet Gynecol 2013;209:229.e1-7. 229-34. 148.e1-6.

518 American Journal of Obstetrics & Gynecology JUNE 2014

www.AJOG.org Obstetrics Expert Reviews
66. Ananth CV, Peltier MR, Chavez MR, 81. Winkvist A, Mogren I, Hogberg U. Familial 96. Moffett A, Loke C. Immunology of placen-
Kirby RS, Getahun D, Vintzileos AM. Recurrence patterns in birth characteristics: impact on indi- tation in eutherian mammals. Nat Rev Immunol
of ischemic placental disease. Obstet Gynecol vidual and population risks. Int J Epidemiol 2006;6:584-94.
2007;110:128-33. 1998;27:248-54. 97. Moffett-King A. Natural killer cells and
67. Bramham K, Briley AL, Seed P, Poston L, 82. Salonen Ros H, Lichtenstein P, pregnancy. Nat Rev Immunol 2002;2:656-63.
Shennan AH, Chappell LC. Adverse maternal Lipworth L, Cnattingius S. Genetic effects on 98. Chazara O, Xiong S, Moffett A. Maternal KIR
and perinatal outcomes in women with previous the liability of developing pre-eclampsia and and fetal HLA-C: a fine balance. J Leukoc Biol
preeclampsia: a prospective study. Am J Obstet gestational hypertension. Am J Med Genet 2011;90:703-16.
Gynecol 2011;204:512.e1-9. 2000;91:256-60. 99. Bulmer JN, Lash GE. Human uterine natural
68. Anselem O, Girard G, Stepanian A, Azria E, 83. Shahabi A, Wilson ML, Lewinger JP, killer cells: a reappraisal. Mol Immunol 2005;42:
Mandelbrot L. Influence of ethnicity on the clin- Goodwin TM, Stern MC, Ingles SA. Genetic 511-21.
ical and biologic expression of pre-eclampsia in admixture and risk of hypertensive disorders of 100. Hiby SE, Walker JJ, O’Shaughnessy KM,
the ECLAXIR study. Int J Gynaecol Obstet pregnancy among Latinas in Los Angeles et al. Combinations of maternal KIR and fetal
2011;115:153-6. County. Epidemiology 2013;24:285-94. HLA-C genes influence the risk of preeclampsia
69. Sibai BM, Gordon T, Thom E, et al. Risk 84. Esplin MS, Fausett MB, Fraser A, et al. and reproductive success. J Exp Med
factors for preeclampsia in healthy nulliparous Paternal and maternal components of the pre- 2004;200:957-65.
women: a prospective multicenter study. The disposition to preeclampsia. N Engl J Med 101. Hiby SE, Regan L, Lo W, Farrell L,
National Institute of Child Health and Human 2001;344:867-72. Carrington M, Moffett A. Association of
Development Network of Maternal-Fetal Medicine 85. Lie RT, Rasmussen S, Brunborg H, maternal killer-cell immunoglobulin-like re-
Units. Am J Obstet Gynecol 1995;172:642-8. Gjessing HK, Lie-Nielsen E, Irgens LM. Fetal and ceptors and parental HLA-C genotypes with
70. Chesley LC, Cooper DW. Genetics of hy- maternal contributions to risk of pre-eclampsia: recurrent miscarriage. Hum Reprod 2008;23:
pertension in pregnancy: possible single gene population based study. BMJ 1998;316: 972-6.
control of pre-eclampsia and eclampsia in the 1343-7. 102. Hiby SE, Apps R, Sharkey AM, et al.
descendants of eclamptic women. Br J Obstet 86. Robillard PY, Dekker GA, Hulsey TC. Maternal activating KIRs protect against human
Gynaecol 1986;93:898-908. Revisiting the epidemiological standard of reproductive failure mediated by fetal HLA-C2.
71. Arngrimsson R, Bjornsson S, Geirsson RT, preeclampsia: primigravidity or primipaternity? J Clin Invest 2010;120:4102-10.
Bjornsson H, Walker JJ, Snaedal G. Genetic and Eur J Obstet Gynecol Reprod Biol 1999;84: 103. Nakimuli A, Chazara O, Farrell L, et al. Killer
familial predisposition to eclampsia and pre- 37-41. cell immunoglobulin-like receptor (KIR) genes
eclampsia in a defined population. Br J Obstet 87. Li DK, Wi S. Changing paternity and the risk and their HLA-C ligands in a Ugandan popula-
Gynaecol 1990;97:762-9. of preeclampsia/eclampsia in the subsequent tion. Immunogenetics 2013;65:765-75.
72. van Dijk M, Oudejans C. (Epi)genetics of pregnancy. Am J Epidemiol 2000;151:57-62. 104. Norman PJ, Hollenbach JA, Nemat-
pregnancy-associated diseases. Front Genet 88. Cincotta RB, Brennecke SP. Family history Gorgani N, et al. Co-evolution of human
2013;4:180. of pre-eclampsia as a predictor for pre- leukocyte antigen (HLA) class I ligands with
73. Mogren I, Hogberg U, Winkvist A, eclampsia in primigravidas. Int J Gynaecol killer-cell immunoglobulin-like receptors (KIR)
Stenlund H. Familial occurrence of preeclamp- Obstet 1998;60:23-7. in a genetically diverse population of
sia. Epidemiology 1999;10:518-22. 89. Chappell S, Morgan L. Searching for genetic sub-Saharan Africans. PLoS Genet 2013;9:
74. Skjaerven R, Vatten LJ, Wilcox AJ, clues to the causes of pre-eclampsia. Clin Sci e1003938.
Ronning T, Irgens LM, Lie RT. Recurrence of (Lond) 2006;110:443-58. 105. Xiong S, Sharkey AM, Kennedy PR, et al.
pre-eclampsia across generations: exploring 90. Zhao L, Bracken MB, DeWan AT. Genome- Maternal uterine NK cell-activating receptor
fetal and maternal genetic components in a wide association study of pre-eclampsia detects KIR2DS1 enhances placentation. J Clin Invest
population based cohort. BMJ 2005;331:877. novel maternal single nucleotide polymorphisms 2013;123:4264-72.
75. Carr DB, Epplein M, Johnson CO, and copy-number variants in subsets of the 106. Rosenberg K, Trevathan W. Birth, obstet-
Easterling TR, Critchlow CW. A sister’s risk: Hyperglycemia and Adverse Pregnancy rics and human evolution. BJOG 2002;109:
family history as a predictor of preeclampsia. Am Outcome (HAPO) study cohort. Ann Hum Genet 1199-206.
J Obstet Gynecol 2005;193:965-72. 2013;77:277-87. 107. Abitbol MM, Chervenah F, Ledger W. Birth
76. Dawson LM, Parfrey PS, Hefferton D, et al. 91. Buurma AJ, Turner RJ, Driessen JH, et al. and human evolution: anatomical and obstetrical
Familial risk of preeclampsia in Newfoundland: a Genetic variants in pre-eclampsia: a meta- mechanics in primates. Westport Bergin and
population-based study. J Am Soc Nephrol analysis. Hum Reprod Update 2013;19: Garvey; 1996.
2002;13:1901-6. 289-303. 108. Cavalli-Sforza LL, Bodmer WF. The ge-
77. Nilsson E, Salonen Ros H, Cnattingius S, 92. Staines-Urias E, Paez MC, Doyle P, et al. netics of human populations. San Francisco:
Lichtenstein P. The importance of genetic and Genetic association studies in pre-eclampsia: W. H. Freeman and Company; 1971:612-3.
environmental effects for pre-eclampsia and systematic meta-analyses and field synopsis. 109. Wittman AB, Wall LL. The evolutionary or-
gestational hypertension: a family study. BJOG Int J Epidemiol 2012;41:1764-75. igins of obstructed labor: bipedalism, enceph-
2004;111:200-6. 93. Goddard KA, Tromp G, Romero R, et al. alization, and the human obstetric dilemma.
78. Plunkett J, Borecki I, Morgan T, Candidate-gene association study of mothers Obstet Gynecol Surv 2007;62:739-48.
Stamilio D, Muglia LJ. Population-based esti- with pre-eclampsia, and their infants, analyzing 110. Thomas J, Paranjothy S. Royal College of
mate of sibling risk for preterm birth, preterm 775 SNPs in 190 genes. Hum Hered 2007;63: Obstetricians and Gynaecologists Clinical Effec-
premature rupture of membranes, placental 1-16. tiveness Support Unit. National Sentinel Cae-
abruption and pre-eclampsia. BMC Genet 94. Gopec Consortium. Disentangling fetal and sarean Section Audit Report. London, UK: RCOG
2008;9:44. maternal susceptibility for pre-eclampsia: a Press; 2001.
79. Toivonen S, Keski-Nisula L, Saarikoski S, British Multicenter candidate-gene study. Am J 111. Cheng YW, Norwitz ER, Caughey AB. The
Heinonen S. Risk of placental abruption in first- Hum Genet 2005;77:127-31. relationship of fetal position and ethnicity with
degree relatives of index patients. Clin Genet 95. Rich-Edwards JW, Fraser A, Lawlor DA, shoulder dystocia and birth injury. Am J Obstet
2004;66:244-6. Catov JM, et al. Pregnancy characteristics and Gynecol 2006;195:856-62.
80. Porter TF, Fraser AM, Hunter CY, Ward RH, women’s future cardiovascular health: an under- 112. Pattinson RC. Pelvimetry for fetal cephalic
Varner MW. The risk of preterm birth across used opportunity to improve women’s health? presentations at term. Cochrane Database Syst
generations. Obstet Gynecol 1997;90:63-7. Epidemiol Rev 2014;36:57-70. Rev 2000:CD000161.

JUNE 2014 American Journal of Obstetrics & Gynecology 519

Expert Reviews Obstetrics www.AJOG.org

113. Handa VL, Lockhart ME, Fielding JR, et al. 118. Loftin R, Chen A, Evans A, Defranco E. Racial differences in the predictive value of the TDx fetal
Racial differences in pelvic anatomy by magnetic differences in gestational age-specific neonatal lung maturity assay. Am J Obstet Gynecol
resonance imaging. Obstet Gynecol 2008;111: morbidity: further evidence for different gestational 1996;175:73-7.
914-20. lengths. Am J Obstet Gynecol 2012;206:259.e1-6. 123. Floros J, Fan R, Diangelo S, Guo X, Wert J,
114. Baragi RV, Delancey JO, Caspari R, 119. Patel RR, Steer P, Doyle P, Little MP, Luo J. Surfactant protein (SP) B associations
Howard DH, Ashton-Miller JA. Differences in Elliott P. Does gestation vary by ethnic group? A and interactions with SP-A in white and black
pelvic floor area between African American and London-based study of over 122,000 preg- subjects with respiratory distress syndrome.
European American women. Am J Obstet nancies with spontaneous onset of labour. Int J Pediatr Int 2001;43:567-76.
Gynecol 2002;187:111-5. Epidemiol 2004;33:107-13. 124. Parham P, Moffett A. Variable NK cell re-
115. Patriquin ML, Steyn M, Loth SR. Metric anal- 120. Balchin I, Whittaker JC, Lamont RF, ceptors and their MHC class I ligands in immu-
ysis of sex differences in South African black and Steer PJ. Timing of planned cesarean delivery nity, reproduction and human evolution. Nat Rev
white pelves. Forensic Sci Int 2005;147:119-27. by racial group. Obstet Gynecol 2008;111: Immunol 2013;13:133-44.
116. Steer PJ. The mechanisms and manage- 659-66. 125. Chi BH, Vwalika B, Killam WP, et al.
ment of normal labour. In: Chamberlain G, 121. Alexander GR, Hulsey TC, Robillard PY, Implementation of the Zambia electronic peri-
Steer P, eds. Turnbull’s obstetrics. London: De Caunes F, Papiernik E. Determinants of natal record system for comprehensive prenatal
Churchill Livingstone; 2001. meconium-stained amniotic fluid in term preg- and delivery care. Int J Gynaecol Obstet
117. Omigbodun AO, Adewuyi A. Duration of nancies. J Perinatol 1994;14:259-63. 2011;113:131-6.
human singleton pregnancies in Ibadan, Nigeria. 122. Berman S, Tanasijevic MJ, Alvarez JG, 126. Kaplan M. Genomics in Africa: avoiding
J Natl Med Assoc 1997;89:617-21. Ludmir J, Lieberman E, Richardson DK. Racial past pitfalls. Cell 2011;147:11-3.

520 American Journal of Obstetrics & Gynecology JUNE 2014

www.AJOG.org Obstetrics Expert Reviews

Appendix physiological transformation of uterine Linkage disequilibrium: differences

Glossary spiral arteries, but there are multiple fetal in a population between the observed
Genetic admixture: phenomenon of and maternal factors that determine the occurrence of some combinations of
interbreeding between members of 2 or exact clinical outcome. alleles or genetic markers and their
more different populations resulting in HLA-C: polymorphic classical HLA expected frequencies under the assump-
the exchange of genes. class I molecules present at the surface of tion of independence. Linkage disequi-
Genome-wide association: a search most nucleated cells and also expressed librium is expected to be the greatest for
for common genetic variants associated by invasive trophoblast cells at the site 2 genes located closely together on a
with a disease. Generally, the association of placentation. HLA-C molecules are chromosome.
of particular single-nucleotide poly- important ligands for KIR receptors. NK cells: large granular lymphocytes
morphism (SNPs), variation in the DNA Killer cell immunoglobulin-like of the innate immune system that play
sequence of a single nucleotide, is ana- receptors (KIR): a family of closely important roles in defense against viral
lyzed in individuals with the disease related genes located in the leukocyte infections and tumors. A specialized
compared with normal matched control receptor complex on chromosome 19. subset of NK cells amass at the site of
subjects. KIR proteins are expressed on the sur- placentation early in gestation.
Great obstetrical syndromes: major face of NK cells. KIR can have an acti- Positive selection: type of natural se-
pregnancy complications such as spon- vating or inhibitory effect on NK cell lection that results in the increase of the
taneous abortion, fetal death, abruptio function. In all populations, 2 types of frequency of the advantageous alleles
placentae, fetal growth failure, pre- KIR haplotypes are found; KIR A hap- over time.
eclampsia, preterm labor, and premature lotypes contain mainly inhibitory KIR, Stabilizing selection: selection favor-
rupture of membranes. These disorders whereas KIR B haplotypes contain extra, ing intermediate values of a character
all share the primary defect of failure of mainly activating, KIR. over the extremes.

JUNE 2014 American Journal of Obstetrics & Gynecology 520.e1