Key words
preterm birth - obesity - pregnancy
Introduction
Maternal overweight and obesity are prenatal risk factors for obstetrical complications
[1], preterm birth [2], neonatal morbidity [3] as well as cognitive and behavioural developmental disorders in children [4]. Paediatric morbidity and mortality as well as child development disorders are significantly
associated with maternal obesity [5], [6], [7], [8]. Particularly in the neurodevelopmental and psychiatric area, it is becoming increasingly
clear that, in children of mothers with an increased body mass index (BMI), there
is a high correlation with childhood cognitive disabilities, attention disorders,
and diseases on the autistic spectrum [6].
The connection and the underlying mechanisms between maternal obesity and the above-mentioned
paediatric morbidity are currently undergoing intensive research. While a main hypothesis
primarily investigates the influence of epigenetic factors [9] which is not intended to be the focus of this work, our working groups investigated
inflammation-associated mechanisms [8]. Here the focus is on the perinatal-neuroepidemiological approach to this issue,
in particular. By analysing large data sets with perinatal and paediatric development
data, we examine the hypothesis whether maternal overweight and obesity in terms of
a chronic inflammatory state is associated with neonatal inflammation which in turn
is associated with an unfavourable development prognosis.
The ELGAN (Extremely Low Gestational Age Newborn) study is a multicentre study which has been supported since 2000 by the National
Institutes of Health (NIH) and whose objective is to research predictors for neonatal
brain damage and neurological-cognitive sequelae in premature infants [10]. The areas of focus are the connection between maternal overweight and obesity and
pregnancy complications, Apgar scores [11] and systemic inflammatory markers [12]. Moreover, a series of analyses directly concern the postulated association between
maternal BMI and neonatal-paediatric consequences [13], [14], [15], [16].
In this overview, our aim is to summarise the work in this area and discuss it critically
on the basis of current literature.
Maternal Weight and Body Mass Index
Maternal Weight and Body Mass Index
Based on the German Obesity Association, obesity is defined as an increase in body
fat beyond the normal range [17]. The body mass index is used internationally to calculate and classify weight classes.
While a BMI between 18.5 and 24.9 kg/m2 is considered to be normal, a BMI between 25 and 29.9 kg/m2 is considered to be overweight and a BMI over 30 kg/m2 is considered to be obese. Obesity, in turn, is divided into degrees of severity,
from I to III (grade I: BMI 30 – 34.9 kg/m2; grade II: 35 – 39.9 kg/m2; grade III > 40 kg/m2) [17].
According to studies by the World Health Organisation (WHO), in 2008, about 1.4 billion
adults worldwide were overweight and at least 500 million adults were obese. The WHO
anticipated 2.3 billion overweight persons in 2015. In the normal population, obesity
has generally doubled in recent decades. In 2014, according to press release 203/2016
from Eurostat, 46.1% of persons living in the EU aged 18 and over were of normal weight,
while slightly more than half of adults (51.6%) were classified as overweight (35.7%
overweight and 15.9% obese). In Germany, 47% of women are affected by overweight and
obesity (BMI > 25 kg/m2). Approximately 29% of all women are overweight (BMI between 25 and 30 kg/m2) and about 18% are obese (BMI over 30 kg/m2) [18]. The prevalence of the combination of overweight and obesity in men between the
ages of 15 and 49 in China between 2010 and 2014 was approximately 22 – 23% [19]. Data from China published in 2002 showed that about 14.7% of the Chinese population
was overweight and another 2.6% were obese [20]. At the time analysed, this represented 184 million people.
The proportion of pregnant women with obesity has also significantly grown. A German
study from 2007 comparing the prevalence of overweight and obesity in pregnant women
between 1980 and 2005 revealed that the number of overweight and obese pregnant women
had tripled, with a disproportionately large increase in severe obesity [21]. In the United Kingdom, nearly 20% of all pregnant women suffer from obesity [22]. In the USA, over 50% of all pregnant women are either overweight or obese [23]. The prevalence of overweight and obesity in pregnant women varies in the different
countries from 1.8 to 25%.
A normal pregnancy and obesity share common characteristics. Mechanisms which are
involved in the pathogenesis of obesity also represent essential parts of the physiological
processes of maternal adaptation to the pregnancy. During pregnancy, weight gain is
normal and desirable. There is a positive correlation between maternal weight gain
in pregnancy and the birth weight of the foetus. In a recommendation from the Institute
of Medicine, pregnant women with a BMI > 30 kg/m2 are recommended a maximum weight gain of 5 to 9 kg [24]. A further reduction in weight gain for obese pregnant women is the subject of controversial
discussion since this is potentially correlated with an increased risk of intrauterine
growth retardation [25].
With regard to nutrition and lifestyle before and during pregnancy, the recommended
actions of the nationwide network “Gesund ins Leben” (Healthy into Life) can be used.
These recommendations address body weight prior to conception, changes in weight during
pregnancy, the energy and nutritional requirements, as well as diet [67], [68].
Maternal Weight and Systemic Inflammation
Maternal Weight and Systemic Inflammation
Obesity [26], [27] as well as pregnancy [28] can lead to a chronic inflammation reaction. Cytokines are protein molecules with
diverse functions. Some cytokines are referred to as growth factors since they initiate
or regulate the proliferation and differentiation of target cells. Other cytokines
play an important role in immunological reactions and inflammatory processes in which
they serve, above all, as signalling molecules between the immune and the nervous
system [29]. The C-reactive protein (CRP) is an acute phase protein which further drives inflammation,
while leptin is an adipokine, which is associated not only with feelings of satiety
and energy homeostasis but also with a pro-inflammation reaction [30].
The systemic response to a pregnancy, which includes the mediators IL-6, CRP and leptin,
among others, was elevated in overweight women prior to conception [12], [31], [32]. In the 4th week of pregnancy, a higher level of CRP could be detected in overweight
women as compared to normal-weight pregnant women [33].
A BMI > 30 kg/m2 represents a significant risk factor for a preterm delivery. Elevated levels of inflammatory
proteins which lead to cervical ripening as well as to myometrial contractions were
assumed to be the cause. As a result of the increased production of adipokines (such
as leptin, for example) by the fat tissue as well as increased secretion of proinflammatory
cytokines, maternal obesity appears to trigger a chronic inflammatory reaction [25].
In 80 blood samples taken from pregnant women during the second trimester, an increase
in MCP-1, a proinflammatory cytokine, produced by macrophages, monocytes and endothelial
cells, as well as an increase in leptin and CRP could be demonstrated in the group
of severely obese pregnant women [12]. The association patterns of the pro- and anti-inflammatory markers with the various
pregnancy characteristics greatly vary [34].
While in the case of spontaneous preterm infants there was no connection between maternal
BMI and increased inflammatory proteins in the children in the ELGAN study, this was
able to be confirmed in the group of deliveries due to maternal or foetal problems
[14]. This so-called effect modification can be explained very well by the fact that
the spontaneous preterm delivery, in contrast to preterm deliveries due to maternal
or foetal indications, is strongly associated with prenatal infection and inflammation.
It can be assumed that in the group of spontaneous preterm deliveries, all women,
thus also those with a normal BMI, have an “inflammatory phenotype”, while this is
not the case in the group of maternal and foetal indications. For this reason, there
may be no perceptible contrast with regard to maternal inflammation due to an elevated
BMI in the case of spontaneously delivered children with or without a neonatal inflammation
reaction. The signal cannot, so to speak, be reliably perceived due to significant
“background noise”.
In the ELGAN study, classification of the phenotypes of the preterm delivery was performed
according to clinical presentation [69]. Of more than 1000 extremely immature preterm infants who were born before the 28th
week of pregnancy, the distribution of the clinical presentation was as follows: Premature
labour 40%, premature rupture of membranes 23%, preeclampsia 18%, placental abruption
11%, cervical insufficiency 5% and foetal indication/growth restriction 3%. In a subsequent
analysis of these data, a significant prevalence contrast between two phenotype clusters
was seen with regard to the neurological outcome: (A) premature labour, rupture of
membranes, cervical insufficiency and placental abruption (3 – 5%) and (B) preeclampsia
or foetal indication (1 – 2%) [70].
The connection between preterm birth phenotype and outcome is mediated, among other
things, by protracted systemic neonatal inflammation [71]. In comparison to preterm infants of the phenotype cluster B, preterm infants of
cluster A have a significantly higher risk for elevated serum concentrations of cytokines
and other inflammatory markers after birth [72]. In statistical cluster analyses of cytokines from placental lysates, it was able
to be shown that placentas after preeclampsia (cluster B) have elevated values for
VEGF (vascular endothelial growth factor) and TGF-beta (transforming growth factor
beta) as well as low inflammatory markers, while about half of the placentas from
cluster A demonstrated an increased inflammation response [73]. The postnatal systemic inflammation reaction correlates with placental infection
and inflammation [74] as well as with an increased risk of neurocognitive developmental disorders at the
age of 10 years [75].
Somatic Consequences for the Child
Somatic Consequences for the Child
Growth
A motherʼs body weight has consequences for the growth pattern of her child. In the
ABCD study, Oostvogels studied more than 3800 mother–child pairs and determined that
during the first years of life, sons as well as daughters of overweight mothers gained
weight and BMI more quickly. These effects are modified by age and gender: Differences
between the observed groups become larger over time and are more pronounced in girls
than in boys [7].
Structural deformities
Maternal overweight and obesity are associated not only with an increased risk of
foetal macrosomia and neonatal mortality [5], but also with structural changes, such as neural tube closure defects, cardiac
anomalies, or orofacial malformations [35]. In a systematic analysis of 18 studies, Stothard et al. [36] showed no fewer than ten such developmental anomalies ([Table 1]). Only in the case of gastroschisis was the likelihood of occurrence reduced.
Table 1 Association between maternal overweight or obesity and structural changes of the
fetus. Results of a systematic analysis of 18 studies based on Stothard et al. [36]. The probability of occurrence was only reduced in the case of gastroschisis..
|
Malformation
|
Odds ratio
|
95% confidence interval
|
|
Source: [36]
|
|
Neural tube defects
|
1.87
|
1.62 – 2.15
|
|
Spina bifida
|
2.24
|
1.86 – 2.69
|
|
Cardiovascular anomalies
|
1.30
|
1.12 – 1.51
|
|
Septal defects
|
1.20
|
1.09 – 1.31
|
|
Cleft palate
|
1.23
|
1.03 – 1.47
|
|
Cleft lip and palate
|
1.20
|
1.03 – 1.40
|
|
Anorectal malformations
|
1.48
|
1.12 – 1.97
|
|
Hydrocephalus
|
1.68
|
1.19 – 2.36
|
|
Hip dysplasia
|
1.34
|
1.03 – 1.73
|
|
Gastroschisis
|
0.17
|
0.10 – 0.30
|
Overweight and obesity
It is possible that the development of obesity is influenced in the prenatal period
and that the maternal weight gain during pregnancy could have an effect on the later
obesity of the child [37]. Such a connection can be explained by the model of so-called “metabolic imprinting”.
This concerns a modification of the intrauterine environment which can have a direct
effect on the BMI of the unborn child [38] and which could thus also represent a risk factor for obesity in adulthood [39], [40]. The hypothesis behind this is that the foetal metabolism is changed due to the
motherʼs malnutrition and hyperglycaemia and the development of obesity is promoted
[40].
Neurological and Cognitive Development
Neurological and Cognitive Development
Overweight and obesity prior to pregnancy are associated with antenatal and peripartum
complications such as gestational diabetes, preeclampsia, pregnancy-induced hypertension
and complications relating to delivery [41], [42]. Moreover, maternal obesity additionally appears to have negative effects on the
newborn [43], such as cognitive deficits [44], [45], autistic developmental disorders [46], [47] or cerebral palsy (CP) [48], [49], [50], [51], [52].
Cerebral palsy
Two comprehensive meta-analyses [53], [54] investigated the relationship between maternal BMI and the risk of cerebral palsy.
Both analyses have a significant connection between maternal overweight or obesity
and the occurrence of cerebral palsy. Maternal overweight and maternal obesity grade
II and grade III were associated with an increased risk of 29, 45 or even 125% [53]. In contrast to this, the data from the ELGAN study did not reveal any increased
risk for CP in children born very prematurely to overweight or obese mothers in comparison
to mothers with a normal weight [55]. As in the effect modification described above through spontaneous versus induced
delivery, the lack of a connection in the case of extremely premature infants could
be due to a greater “inflammatory background noise” as compared to infants born at
term.
Neurocognitive development
In a small, monocentre study of 62 maternal/child pairs in whom delivery occurred
before the end of the 31st week of pregnancy, maternal obesity was associated with
a positive autism screening and low speech development score [56]. In the ELGAN study, the pre-pregnancy heights and weights of the mothers of 852
children born prematurely were collected and analysed in a multinomial logistic regression
model. It showed that, compared to newborns of mothers with a normal BMI, newborns
of obese but not of overweight mothers had a greater likelihood of reaching Bayley
Scales indices more than three standard deviations below the reference range (mental
scale: OR = 2.1; 95% CI: 1.3, 3.5; motor scale: OR = 1.7; 95% CI: 1.1, 2.7) [13]. This association was even greater in newborns who did not demonstrably have any
intermittent or longer-lasting systemic inflammatory marker profiles. Maternal obesity
accordingly appears to be associated with an increased risk of impaired development
of the newborn.
At the age of 10 years, an increased risk for decreased scores in the verbal ability
scale II, IQ measurements for processing speed and fine motor control (developmental
neuropsychological assessment II) as well as for pronunciation and spelling (Wechsler
individual achievement test-III) were seen in an analysis of 535 children from the
ELGAN study [15]. Children of mothers who gained excessive weight during pregnancy had an increased
risk of low scores in the area of linguistic expression. However, children of mothers
without adequate weight gain also had an increased risk of low scores in the areas
of linguistic expression and reading ability. Physiological weight gain during pregnancy
thus appears to have a protective influence on the neurocognitive development of the
child.
Inflammation as a Pathomechanism
Inflammation as a Pathomechanism
A possible pathomechanism for the connections between maternal weight and neuropaediatric
outcome listed is perinatal inflammation [8].
The scenario of infection during pregnancy, inflammation reaction in the mother and
child, preterm birth and neonatal brain damage have already been postulated for more
than 20 years [57], [58] and have also been extensively documented in the meantime [59], [60]. This results in the following syllogism:
-
Maternal overweight and obesity are associated with preterm delivery and maternal-foetal
inflammation;
-
Maternal overweight and obesity are associated with perinatal brain damage and later
developmental disorders;
-
Preterm birth and inflammation are associated with perinatal brain damage and later
developmental disorders;
-
Maternal overweight and obesity could lead to developmental disorders in preterm infants
via a systemic inflammation reaction.
Which potential role the systemic inflammation reaction plays in the connection between
maternal overweight or obesity and the developmental disorders in preterm infants
remains to be evaluated in more detail. Maternal overweight or obesity can, just like
the pregnancy itself [28], contribute to a chronic inflammation reaction in the mother via cytokines such
as CRP, IL-6, and/or leptin [26], [27]. This maternal systemic inflammation can lead to direct foetal inflammation with
damage to the childʼs brain [61]. Children whose mothers had high levels of the equally proinflammatory mediators
TNF-a [62] or IL-8 [63] have an increased risk for developing schizophrenia. Maternal obesity appears to
contribute to prolonged systemic inflammation in the newborn [14], which in turn represents a significant developmental risk for the child born preterm.
Summary
In summary, it can be concluded that there are multiple valid indications for a connection
between overweight or obesity in mothers and a broad spectrum of developmental disorders
in their children. We primarily discussed the results of the ELGAN study in which
exclusively preterm infants with a gestational age of < 28 weeks of pregnancy were
recruited. These results may therefore not be able to be fully applied to children
with a gestational age of > 28 weeks of pregnancy.
Moreover, there are plausible reasons for explaining these connections through systemic
foetal and neonatal inflammation reactions which are a significant focus of the content
of the ELGAN study. In this investigation, we accordingly focused on these pathomechanisms.
There are of course multiple other possibilities, such as a folate deficiency in obesity
during pregnancy as a potential risk factor in newborns [64].
Weight gain during pregnancy is normally desirable. Nonetheless, an intervention for
risk reduction in the case of overweight and obesity in pregnancy should be discussed
for the reasons discussed here [65]. The evidence with regard to possible efficacy, for example, in the form of dietary
advice during pregnancy for the prevention of gestational diabetes, remains unclear.
To date there are no neonatal data and development data [66]. Corresponding studies would therefore benefit in particular from cooperation between
obstetricians and developmental paediatricians.
