Diabetes Care Volume 37, November 2014
2996
Maternal Hyperglycemia During
Pregnancy Predicts Adiposity of
the Offspring
Ai Kubo,1 Assiamira Ferrara,1
Gayle C. Windham,2 Louise C. Greenspan,3
Julianna Deardorff,4 Robert A. Hiatt,5
Charles P. Quesenberry Jr.,1
Cecile Laurent,1 Anousheh S. Mirabedi,1
and Lawrence H. Kushi1
EPIDEMIOLOGY/HEALTH SERVICES RESEARCH
Diabetes Care 2014;37:2996–3002 | DOI: 10.2337/dc14-1438
OBJECTIVE
To investigate associations between maternal pregnancy hyperglycemia, gestational diabetes mellitus (GDM), and offspring adiposity.
RESEARCH DESIGN AND METHODS
We evaluated these associations in a longitudinal study of 421 mother-daughter
pairs at Kaiser Permanente Northern California. Maternal pregnancy glucose values were obtained from maternal medical records. Outcomes included three
measures of girls’ adiposity, measured annually: 1) ‡85th age-specific percentile
for BMI; 2) percent body fat (%BF); and 3) waist-to-height ratio (WHR).
RESULTS
Adjusting for maternal age at delivery, race/ethnicity, pregravid BMI, girl’s age,
and girl’s age at onset of puberty, having a mother with GDM increased a girl’s risk
of having a BMI ‡85th percentile or having %BF or WHR in the highest quartile
(Q4), compared with those in the lowest quintile of blood glucose (odds ratio [OR]
3.56 [95% CI 1.28–9.92]; OR 3.13 [95% CI 1.08–9.09]; and OR 2.80 [95% CI 1.00–
7.84], respectively). There was a significant interaction between the presence of
GDM and pregravid BMI; girls whose mothers had both risk factors had the highest
odds of having a BMI ‡85th percentile (OR 5.56 [95%CI 1.70–18.2]; Q4 %BF, OR
6.04 [95%CI 1.76–20.7]; and Q4 WHR, OR 3.60 [95%CI 1.35–9.58]). Similar, although weaker, associations were found in the association between hyperglycemia and offspring adiposity.
CONCLUSIONS
Girls who were exposed to maternal GDM or hyperglycemia in utero are at higher
risk of childhood adiposity; risk increases if the mother is overweight or obese.
Screening and intervention for this high-risk group is warranted to slow the intergenerational transmission of obesity and its sequelae.
By age 2 years, almost one-third of U.S. children are overweight or obese (1), and
childhood obesity strongly predicts adult obesity and chronic diseases (2). The
perinatal period offers a critical opportunity for obesity prevention: the programming of obesity starts very early, even in utero, where gestational and perinatal
factors affect the offspring’s obesity trajectory and metabolic imprinting (3). The
White House Task Force on Childhood Obesity (4) and the Surgeon General (5)
recommend promoting effective perinatal interventions in an effort to interrupt
the intergenerational cycle of obesity. Early prevention is critical because once
established, reversal of obesity is often inefficient, ineffective, and costly (6,7).
1
Kaiser Permanente Division of Research, Oakland, CA
2
California Department of Public Health, Richmond,
CA
3
Kaiser Permanente San Francisco Medical Center, San Francisco, CA
4
University of California, Berkeley, Berkeley, CA
5
University of California, San Francisco, San
Francisco, CA
Corresponding author: Ai Kubo, ai.kubo@kp.org.
Received 10 June 2014 and accepted 31 July
2014.
This article contains Supplementary Data online
at http://care.diabetesjournals.org/lookup/
suppl/doi:10.2337/dc14-1438/-/DC1.
© 2014 by the American Diabetes Association.
Readers may use this article as long as the work
is properly cited, the use is educational and not
for profit, and the work is not altered.
care.diabetesjournals.org
It has been hypothesized that fetal
overnutrition (i.e., excess maternal fuels
including higher blood glucose levels)
causes permanent fetal changes, leading to increased risk of obesity and insulin resistance later in life (8,9). Several
previous studies provide evidence that
children born with intrauterine exposure to type 2 diabetes or gestational
diabetes mellitus (GDM) are at increased
risk for childhood obesity and type 2 diabetes later in life and that these risks
are higher than would be predicted
from genetics alone (10–17). However,
results have been inconsistent, and the
studies have had several limitations
(18). For instance, most previous studies used offspring BMI alone as a measure of obesity. However, it is critical to
determine the percent of body fat that
accounts for total BMI, as BMI is not an
accurate indicator of body fat (i.e., individuals with greater muscle mass will
have higher BMIs). Because body fat
is a tissue with endocrine and immune
functions (19), more accurately assessing it is critical, as a higher percentage
may predict metabolic disorders such
as insulin resistance and metabolic syndrome (20). Also, abdominal fat is more
active metabolically than peripheral fat
and is more strongly associated with the
aforementioned adult chronic conditions and cardiovascular disease (21).
Thus, a better understanding of the
risk factors associated with development of abdominal obesity and greater
adiposity (% fat) may have important
public health implications. Second,
most previous studies focused on the
effect of diabetes or GDM; little is
known regarding the effects of subclinical maternal pregnancy hyperglycemia
on childhood adiposity. Lastly, few prior
studies have included important covariates such as maternal pregravid obesity,
child birth weight, and pubertal development data.
Hyperglycemia during pregnancy is
modifiable, treatable, and preventable.
To further our efforts to design highyield targets for intervention, we evaluated the association between maternal
glycemic level during pregnancy and
measures of adiposity in female offspring, including BMI, abdominal obesity, and percent of total body fat. The
analyses were conducted in a longitudinal study of multiethnic, adolescent girls
in northern California drawn from the
Kubo and Associates
membership of a large prepaid health
plan with demographic characteristics
similar to the general population (22).
RESEARCH DESIGN AND METHODS
This study was carried out as part of
the Puberty Studies of the National Institute of Environmental Health Sciences/
National Cancer Institute Breast Cancer
and the Environment Research Program,
three cooperative studies examining determinants of early puberty in prospective cohort studies (23). The present
analysis used data from one of these epidemiologic projects based in the San
Francisco Bay area, the Cohort Study of
Young Girls’ Nutrition, Environment, and
Transitions (CYGNET). Written informed
consent from the parents, and assent
from the children, was obtained. The
study protocol was approved by the Kaiser
Permanente Institutional Review Board.
Participants and Procedure
This study recruited 444 girls and their
caregivers (96% biological mothers),
currently members of Kaiser Permanente Northern California (KPNC). Details of the study protocol have been
described previously (24). Briefly, girls
were 6–8 years old at baseline and ethnically diverse. At each annual clinic visit
(mean follow-up in this analysis: 3.8
years [range 2–6 years]), anthropometric measurements and Tanner staging
for breast and pubic hair development
were assessed. Information was collected in various phases through interviews conducted with caregivers and as
well with the girls themselves when they
were old enough (age 10–12 y) to provide their own information.
Measurements
Exposure Variables
Maternal Pregnancy Glucose Levels.
Each
girl’s medical record was linked to her
mother’s medical record. At KPNC, 95%
of pregnant women undergo the recommended 50-g, 1-h glucose challenge test
for GDM screening (25) during gestational weeks 24–28 (hereafter referred
to as the screening test). All plasma glucose measurements were performed using the hexokinase method by the
regional laboratory of KPNC. This laboratory participates in the College of
American Pathologists’ accreditation
and monitoring program. Screening
test results were categorized into quintiles. The ranges of plasma glucose
levels for each of the quintile categories
were: #90, 90–103, 104–118, 119–140,
and $141 mg/dL. The highest quintile
corresponded with the cutoff point
used at KPNC to define abnormal
screening (1-h glucose $140 mg/dL),
and therefore women in this group underwent the diagnostic, 100-g, 3-h oral
glucose tolerance test.
GDM was defined according to the
Carpenter and Coustan thresholds (26)
for the diagnostic test.
Maternal pregravid BMI was calculated as (weight in kg)/(height in meters
squared) from self-reported weight and
height obtained from the CYGNET Study
questionnaire. We categorized the women
as normal weight (BMI ,25.0 kg/m2),
overweight (BMI 25.0 to #30.0 kg/m2),
and obese (BMI $30.0 kg/m2).
Outcomes
Offspring BMI. The annual study clinic visit
included several anthropometric measurements. Girls’ height was measured to
the nearest 0.1 cm, using a mounted wall
stadiometer, with the participant in stocking feet and head in the neutral position.
Weight was measured without shoes and
in light clothing and was rounded to the
nearest 0.5 kg. BMI percentile and z-score
were calculated in comparison with the
appropriate age- (and sex-) specific Centers
for Disease Control and Prevention year
2000 standard population distribution (27).
Waist-to-Height Ratio. Waist circumference
was measured twice at the umbilicus,
and the difference between the two
measurements was calculated. If the difference was .1 cm, then a third measurement was taken, and the average for
these two or three measurements was
used for analysis. Hip circumference
was measured at the widest part of the
hips following a similar protocol. Waistto-height ratio (WHR = waist/height) was
calculated as a measure of abdominal
obesity.
Percent of Total Body Fat. Percent of total
body fat was estimated from bioelectrical
impedance analysis from the Tanita scale,
which uses a small electrical current from
foot to foot to estimate adiposity.
Covariates
Girls’ demographic information, baseline dietary (percent kcal from fat and
total daily energy intake), and physical
activity information (using average daily
metabolic equivalent), as well as maternal
2997
2998
Diabetes Care Volume 37, November 2014
Pregnancy Hyperglycemia and Daughter’s Obesity
demographics (income and education)
and smoking status during pregnancy,
were obtained from the CYGNET questionnaires. Girl’s ethnicity was reported by the
parent or caregiver and categorized as:
African American or black, Latina, Asian,
or non-Hispanic white. Pubertal onset was
assessed using Tanner staging, an established five-stage classification scheme (28),
performed by trained personnel (L.C.G.).
Statistical Analyses
Data were analyzed using SAS statistical
software version 9.3 (SAS Institute, Cary,
NC). The demographic and lifestyle characteristics of mothers and daughters
were compared among quintile categories of mother’s pregnancy glucose levels
using ANOVA for continuous variables
and x2 test for categorical variables. To
estimate the association between maternal pregnancy plasma glucose levels
on daughters’ obesity for each year of
follow-up, we used the generalized estimating equation method, which takes into
account correlations among intraindividual outcomes in repeated measuresdin
this case, repeat assessments of adiposity
over 6 years of follow-up.
We defined high adiposity using several measures: $85th percentile BMI
for age, fourth (highest) quartile (Q4)
for percent body fat (%BF), and Q4 for
WHR. Maternal pregnancy glucose levels were categorized by quintiles of the
glucose value, and those in the second
to fifth quintile were compared with
women in the first (lowest) quintile category (referent). Women in the fifth
quintile were categorized into GDM versus no GDM groups. The main models
were adjusted for race/ethnicity, maternal pregravid BMI (continuous), girls’
age (months), maternal age at delivery
(years), and age at onset of puberty
(months; i.e., Tanner stages 2+ for
breast and pubic hair development).
Second, we investigated whether
there was an interaction between maternal pregravid obesity and maternal
pregnancy glucose levels on offspring
adiposity. For this analysis, girls with
mothers with normal pregravid BMI
and lowest quintile of pregnancy glucose level (referent) were compared
with those with one or two of these
risk factors (higher glucose and/or pregravid obesity).
As a supplementary analysis, we examined the role of birth weight on the
pathway between maternal glucose or
obesity and offspring obesity. We included birth weight (as continuous
as well as categorical: macrosomia
.4,000 g; normal weight; and low birth
weight ,2,500 g) in the model to see if
the main association was attenuated.
RESULTS
Participant Characteristics
Of the 444 participants, we excluded 23
mother-daughter pairs who were missing maternal pregnancy glucose values,
resulting in 421 pairs with baseline information included in the analysis. Table
1 presents the characteristics of the
mothers. There were no significant differences for education, income, BMI, or
smoking during pregnancy by glucose
levels. A total of 27 women had GDM
(26 in the fifth quintile and 1 in the
fourth quintile). There was a nonsignificant trend for older maternal age at delivery being associated with higher
glucose values.
Table 2 presents the characteristics of
the girls at baseline and at birth. Older
gestational age at birth and higher total
daily energy and fat intake among girls
was associated with mothers who had
lower glucose values during pregnancy.
Primary Analyses
Association Between Maternal Pregnancy
Glucose Levels and Girls’ Adiposity
Table 3 presents associations between
maternal pregnancy glucose levels with
girls’ adiposity, controlling for race/
ethnicity, maternal pregravid BMI,
girls’ age, Tanner stages, and maternal
age at delivery. While accounting for
within-subject correlation among the repeated measures, girls whose mothers
had the highest quintile of pregnancy
glucose were at significantly increased
odds of having BMI $85th percentile
(odds ratio [OR] 2.28 [95% CI 1.08–
4.84]), being in the upper quartile (Q4)
of %BF (OR 2.51 [95% CI 1.16–5.40]), and
Q4 of WHR (OR = 2.48 [95% CI 1.17–
5.22]) compared with girls whose mother
were in the lowest glucose level (Q1). Test
for trends across glucose levels were
statistically significant for %BF and
were borderline for WHR.
Presence of GDM
When women in the highest quintile
of glucose were stratified by presence/
absence of GDM (Table 3), the association
with girls’ adiposity measures was consistently stronger among those with GDM
compared with those without (OR 3.56
[95% CI 1.28–9.92] for BMI; OR 3.13
[95% CI 1.08–9.09] for %BF; OR 2.80
[95% CI 1.00–7.84] for WHR).
Combined Effects of Pregnancy Blood
Glucose Level and Maternal Obesity on
Offspring Adiposity. Given the known as-
sociation between maternal pregravid
obesity and offspring obesity, we evaluated whether there was a combined effect of two risk factors: high pregnancy
blood glucose level (being in the fifth
quintile) and maternal pregravid overweight (BMI $25 kg/m2) (Supplementary Fig. 1). Having both risk factors
increased the risk of girls’ adiposity by
approximately fourfold, compared with
those with neither risk factor. For instance, if a mother had a pregnancy glucose level in the fifth quintile and
pregravid BMI $25.0 kg/m2, the odds
of her daughter being in the BMI category $85 percentile (OR 3.73 [95% CI
1.89–7.37]), Q4 %BF (OR 3.78 [95% CI
1.87–7.66]), and Q4 WHR (OR 3.93
[95% CI 2.02–7.66]) were substantially
higher. Having only a single risk factor
did not result in a significant increase in
the risk of girls’ obesity.
We further tested for an interaction
between presence of GDM and pregravid
BMI and analyzed the combined effect on girl’s adiposity (Table 4). Girls
whose mothers had both of these risk
factors had the highest odds of having
BMI $85th percentile (OR 5.56 [95% CI
1.70–18.2]), high (Q4) %BF (OR 6.04
[95% CI 1.76–20.7]), and high (Q4)
WHR (OR 3.60 [95% CI 1.35–9.58]). Having pregravid BMI $25 absent GDM also
increased the risk of girl’s obesity,
though at much smaller magnitude
(60–70% increase compared with referent). Having GDM with normal BMI was
not associated with offspring’s adiposity, although the sample size was small
(n = 11), and we had insufficient power
to assess these associations.
Supplemental Analyses
Mediation by Birth Weight
To evaluate whether the observed associations were mediated by girl’s birth
weight, we conducted a secondary analysis including child birth weight. In this
population, inclusion of birth weight either as a continuous or categorical variable did not change the effect estimates
care.diabetesjournals.org
Kubo and Associates
Table 1—Demographic characteristics of the mothers by pregnancy glucose levels: CYGNET Study, 2005–2006
Glucose
Q1
Q2
Q3
Q4
Q5
#90
91–104
105–119
120–141
.141
84
82
86
84
85
Age at delivery (years)
32.0 (5.7)
31.4 (6.6)
33.1 (5.9)
33.1 (5.6)
33.5 (5.1)
Education
High school or less
Some college
College/university
Postgraduate
12 (16%)
25 (22%)
31 (23%)
15 (17%)
17 (23%)
21 (18%)
29 (22%)
13 (15%)
13 (17%)
20 (17%)
31 (23%)
20 (23%)
16 (21%)
22 (19%)
23 (17%)
22 (25%)
17 (23%)
28 (24%)
20 (15%)
19 (21%)
Income
,50,000/year
$50,000/year
12 (14%)
71 (22%)
23 (27%)
56 (17%)
17 (20%)
67 (20%)
16 (19%)
68 (21%)
16 (19%)
68 (21%)
BMI
Normal (BMI ,25 kg/m2)
Overweight (25 # BMI , 30 kg/m2)
Obese (BMI $30 kg/m2)
49 (23%)
18 (17%)
11 (16%)
39 (18%)
24 (23%)
11 (16%)
49 (23%)
20 (19%)
13 (19%)
39 (18%)
21 (20%)
16 (24%)
38 (18%)
21 (20%)
17 (25%)
Smoking during pregnancy
4 (20%)
4 (20%)
3 (15%)
3 (15%)
6 (30%)
Range (mg/dL)
n
P value*
0.09
0.62
0.22
0.67
0.82
2
Data are N (%) or mean (SD). *ANOVA for continuous variables and x test for categorical variables.
substantially. For the highest quintile of
maternal glucose compared with the
lowest quintile, the OR for predicting
offspring having a BMI .85th percentile
was 2.15 (95% CI 1.02–4.52) including
birth weight (categorical) did not differ
substantially from our prior result (OR
2.28 [95% CI 1.08–4.84]). This suggests
that birth weight does not fully mediate
the association between maternal glucose levels and offspring adiposity.
CONCLUSIONS
In our multiethnic group of motherdaughter pairs, we observed that
maternal pregnancy hyperglycemia,
whether just below the diagnostic
threshold of GDM or as GDM, were
both associated with increased risk of
childhood obesity, findings that held independent of maternal age at delivery,
race/ethnicity, pregravid BMI, girls’ age,
and age at onset of puberty. In addition,
we found that the risk of childhood
obesity was highest among offspring
of mothers with GDM and pregravid
obesity.
Table 2—Baseline characteristics of the girls by maternal pregnancy glucose levels, CYGNET Study, 2005–2006
Glucose
Q1
Q2
Q3
Q4
Q5
#90
91–104
105–119
120–141
.141
84
82
86
84
85
Age (years)
6
7
8
19 (19%)
62 (20%)
3 (50%)
22 (22%)
59 (19%)
1 (17%)
16 (16%)
70 (22%)
0 (0%)
21 (21%)
62 (20%)
1 (17%)
21 (21%)
63 (20%)
1 (17%)
Race/ethnicity
White
Asian
Latina
African American
41 (23%)
6 (12%)
12 (12%)
25 (29%)
35 (19%)
9 (17%)
19 (19%)
19 (22%)
35 (19%)
14 (27%)
19 (19%)
18 (21%)
37 (20%)
10 (19%)
24 (24%)
13 (15%)
33 (18%)
13 (25%)
27 (27%)
12 (14%)
Tanner stage
Breast $2
Pubic hair $2
6 (20%)
7 (23%)
5 (17%)
8 (26%)
7 (23%)
6 (19%)
8 (27%)
6 (19%)
4 (13%)
4 (13%)
BMI
Normal (BMI ,85th percentile)
Overweight (85th # BMI , 95th)
Obese (BMI $95th)
67 (22%)
8 (13%)
9 (15%)
56 (19%)
11 (18%)
15 (25%)
58 (19%)
17 (28%)
11 (18%)
64 (21%)
12 (20%)
8 (13%)
56 (19%)
12 (20%)
17 (28%)
Range (mg/dL)
N
P value*
0.64
0.13
0.79
0.83
0.26
%BF
17.9 (8.2)
19.2 (8.9)
19.1 (8.3)
18.9 (8.7)
20.5 (10.0)
0.46
WHR
0.48 (0.047)
0.49 (0.057)
0.49 (0.05)
0.48 (0.048)
0.50 (0.059)
0.08
Total energy intake (kcal/day)
1,623 (275)
1,572 (322)
1,627 (468)
1,533 (308)
1,493 (313)
0.05
Total fat intake (g/day)
58.3 (13.9)
53.0 (12.6)
55.6 (18.9)
52.5 (14.4)
51.5 (14.4)
0.03
1.9 (1.3)
1.9 (1.3)
1.9 (1.3)
1.8 (1.3)
1.6 (1.4)
0.68
3,305 (499)
39.3 (1.9)
3,350 (560)
39.5 (2.0)
3,425 (528)
39.4 (1.6)
3,430 (532)
39.1 (1.6)
3,346 (695)
38.6 (2.6)
0.54
0.06
Physical activity: LN METs
Birth weight (g)
Gestational age at birth (weeks)
Data are N (%) or mean (SD). LN, natural log; MET, metabolic equivalent. *ANOVA for continuous variables and x2 test for categorical variables.
2999
3000
Diabetes Care Volume 37, November 2014
Pregnancy Hyperglycemia and Daughter’s Obesity
Table 3—Associations between maternal pregnancy glucose levels and offspring girls’ BMI, %BF, and WHR: CYGNET Study,
2005–2012*
Offspring adiposity measures
BMI $85th percentile
Quintile of maternal glucose (mg/dL)
N
OR
Q1: ,90
75
1.00
Q4 %BF
95% CI
Reference
N
OR
71
1.00
Q4 WHR
95% CI
Reference
N
OR
75
1.00
95% CI
Reference
Q2: 91–104
73
1.85
0.86
3.97
62
1.35
0.60
3.05
73
1.68
0.77
3.65
Q3: 105–119
76
2.08
0.97
4.46
65
2.18
0.99
4.79
76
1.76
0.82
3.79
Q4: 120–141
74
1.26
0.55
2.89
67
1.66
0.76
3.60
74
1.22
0.54
2.77
Q5: .141
Test for trend (P value)
72
2.28
0.13
1.08
4.84
63
2.51
0.03
1.16
5.40
72
2.48
0.07
1.17
5.22
Q5: without GDM
49
1.81
0.79
4.15
46
1.88
0.80
4.42
49
2.11
0.93
4.78
Q5: with GDM
19
3.56
1.28
9.92
13
3.13
1.08
9.09
19
2.80
1.00
7.84
*Generalized estimating equation models adjusted for race/ethnicity, maternal pregravid BMI, girls’ age, Tanner stages, and maternal age at
delivery.
To our knowledge, this is one of the
first studies to examine the effect of
pregnancy hyperglycemia and GDM on
offspring’s adiposity beyond the measure of general obesity (using the proxy
BMI), adjusting for maternal pregravid
BMI. This finding corroborates previous
well-designed studies suggesting that
exposure to maternal GDM or type 2 diabetes in utero may increase the risk of
offspring adiposity in a large group (n .
7,000) of German children (29), in multiethnic children age 6–13 in Colorado
(10), and among younger children in
eastern Massachusetts (30). In these latter studies, exposure to maternal GDM
was associated with higher waist circumference (10) and adiposity (30)
(measured by skinfold thickness), but
not BMI after adjustment for important
covariates including maternal pregravid
BMI. Our results corroborate this finding
as the dose-effect associations between
pregnancy glucose levels and adiposity
outcomes were stronger for percent fat
and WHR than BMI (Table 3). %BF and
abdominal (visceral) obesity are better
predictors of adult chronic conditions
such as insulin resistance, metabolic
syndrome, type 2 diabetes, and cardiovascular disease than BMI (20,21,31).
Our results thus suggest that exposure
to hyperglycemia or GDM in utero increases the risk of metabolic dysregulation in the offspring, manifested as
greater adiposity.
Second, most prior studies evaluated
the association of childhood obesity
with diagnosed or self-reported GDM
or type 2 diabetes (10–16), and little is
known regarding the effect of maternal
pregnancy glucose levels below the diagnostic threshold of GDM. Our results
indicate that pregnancy glucose levels
.140 mg/dL in the screening test, which
coincides with the cutoff point for the
diagnostic test for GDM ($140 mg/dL),
is significantly associated with adiposity
of offspring girls. Among the women in
the highest quintile, only one-third (n =
26) met the criteria of GDM. While elevated maternal glucose levels without
GDM carried an increased risk of offspring adiposity, our supplemental analyses comparing the results between
women with and without GDM demonstrated that the associations were consistently stronger among women with
GDM (Table 3). Furthermore, we also
found that female offspring of women
with pregravid obesity and GDM had
three to six times higher risk of obesity
compared with the referent without either of the risk factors. This finding highlights the importance of identifying
these high-risk women and intervening
with treatment and lifestyle modification. During the pregnancy “teachable
moment,” GDM or overweight women
may be alerted to the increased risk of
their offspring being overweight, which
could potentially encourage the adoption of preventive behaviors, leading to
upstream intervention of childhood
obesity.
Lastly, we found that the association
between maternal pregnancy glucose
and offspring adiposity was not mediated by birth weight. This results supports the findings from long-term Pima
Indian study (32) and suggests that
the long-term consequences of fetal
overnutrition are not fully explained by
increased fetal growth. Similarly, adjustment for girls’ diet (total fat and energy
intake) or physical activity did not
change the associations (data not
shown). Exposure to high levels of glucose in utero, therefore may predispose
the offspring to altered metabolic patterns, affecting long-term regulation of
energy balance. These exposures may influence the development of hypothalamic
Table 4—Association of maternal GDM and pregravid BMI on offspring adiposity, CYGNET Study, 2005–2012
BMI $85th percentile
N
OR
Q4 %BF
95% CI
N
OR
Q4 WHR
95% CI
N
OR
95% CI
BMI ,25 and no GDM
202
1.0
Reference
178
1.0
Reference
202
1.0
BMI ,25 and GDM
11
0.58
0.15
2.27
7
0.39
0.08
1.99
11
1.40
Reference
0.43
4.60
BMI $25 and no GDM
159
1.71
1.08
2.72
144
1.59
1.00
2.55
159
1.78
1.09
2.91
BMI $25 and GDM
P value for interaction
11
5.56
0.03
1.70
18.2
9
6.04
0.01
1.76
20.7
11
3.60
0.01
1.35
9.58
care.diabetesjournals.org
circuits that regulate body weight, as
well as endocrine pancreatic function,
changes in the proportion of lean versus fat body mass, and other cycles of
metabolic programming (33).
While our findings represent important additions to the evidence base,
they should be interpreted with caution.
First, maternal pregravid weight and
height were self-reported. However, because of the unique availability of the
linkage of child medical records to maternal records at KPNC, we were able to
compare the self-reported data with
those using measured height and weight
during the pregnancy in the electronic
medical record (measured at alpha fetoprotein test). The results were similar
(data not shown), making it unlikely
that self-report per se biased the results. Second, since this is an observational study, we cannot preclude
potential confounding by unmeasured
factors that could explain the observed
association. The availability of lifestyle,
demographic, and clinical data, such as
girls’ diet, physical activity, Tanner stage,
and maternal pregravid BMI, collected
in-person from the study enabled us to
investigate the effect of maternal glucose levels independent of these factors
on offspring adiposity. However, these
factors are not measured with complete
accuracy, and it is possible that there are
other unmeasured psychosocial, genetic, and environmental factors that
may explain some of the association.
Lastly, only girls were included in this
study. It is possible that there is effect
modification by sex in the observed
associations.
In conclusion, in this ethnically diverse sample of mother-daughter pairs,
subclinical hyperglycemia and GDM
were both found to be important predictors of offspring adiposity in girls,
independent of maternal pregravid obesity. The concept of windows of susceptibility over the life course provides an
important conceptual framework for
understanding how prenatal exposures
may influence the health of offspring
and the cycle may continue over generations. Our study suggests that exposures during the intrauterine period
affect the offspring’s obesity trajectory,
which will subsequently shape health
outcomes of the girls later in life. Because of this, as Gillman and Ludwig
(7) state in a recent perspective article,
Kubo and Associates
it is imperative that research on childhood obesity move more upstream, focusing on pregnant women or even
women who are planning on becoming
pregnant in order to improve lifelong
health trajectories of the women and
the offspring. Ongoing intervention
studies targeting pregnant or preconceptual women (34) have promising
clinical and public health implications
to slow the ever-increasing rate of obesity and diabetes in the U.S.
Acknowledgments. The authors thank CYGNET
participants, caregivers, and research staff
and Amy Markowitz, University of California,
San Francisco, for help in preparation of the
manuscript.
Funding. This study was supported by grants
U01-ES-012801 and U01-ES-019435 from the
National Institute of Environmental Health
Sciences and the National Cancer Institute and
UL1-RR-024131 from the National Center for
Research Resources. Support was also provided
by the CDPH and Avon Foundation. A.K. is
supported by Career Development Award K07CA-166143-01A1 from the National Cancer Institute and National Institutes of Health Office
of Research on Women’s Health and by the
National Center for Advancing Translational Sciences of the National Institutes of Health (KL2TR-000143).
Duality of Interest. No potential conflicts of
interest relevant to this article were reported.
Author Contributions. A.K. obtained funding,
researched data, and wrote the manuscript. A.F.
reviewed and edited the manuscript and contributed to discussion. G.C.W., L.C.G., J.D., R.A.H., and
L.H.K. designed the study, obtained data, and
reviewed and edited the manuscript. C.P.Q.
researched data. C.L. analyzed data and reviewed
and edited the manuscript. A.S.M. collected data.
A.K. is the guarantor of this work and, as such, had
full access to all the data in the study and takes
responsibility for the integrity of the data and the
accuracy of the data analysis.
References
1. Wojcicki JM, Heyman MB. Let’s Moved
childhood obesity prevention from pregnancy
and infancy onward. N Engl J Med 2010;362:
1457–1459
2. Biro FM, Wien M. Childhood obesity and
adult morbidities. Am J Clin Nutr 2010;91:
1499S–1505S
3. Ong KK. Early determinants of obesity. Endocr Dev 2010;19:53–61
4. Solving the Problem of Childhood Obesity
Within a Generation: White House Task Force
on Childhood Obesity Report to the President.
Washington, DC, U.S. Govt. Printing Office,
2010. Available from http://www.letsmove
.gov/sites/letsmove.gov/files/TaskForce_on_
Childhood_Obesity_May2010_FullReport
.pdf. Accessed 6 August 2014
5. U.S. Department of Health and Human Services. The Surgeon General’s Vision for a Healthy
and Fit Nation. Rockville, MD, U.S. Department
of Health and Human Services, 2010. Available
from http://www.surgeongeneral.gov/initiatives/
healthy-fit-nation/obesityvision2010.pdf.
Accessed 6 August 2014
6. Oken E, Gillman MW. Fetal origins of obesity.
Obes Res 2003;11:496–506
7. Gillman MW, Ludwig DS. How early should
obesity prevention start? N Engl J Med 2013;
369:2173–2175
8. Freinkel N. Banting Lecture 1980. Of pregnancy and progeny. Diabetes 1980;29:1023–
1035
9. Dabelea D. The predisposition to obesity and
diabetes in offspring of diabetic mothers. Diabetes Care 2007;30(Suppl. 2):S169–S174
10. Crume TL, Ogden L, West NA, et al. Association of exposure to diabetes in utero with adiposity and fat distribution in a multiethnic
population of youth: the Exploring Perinatal
Outcomes among Children (EPOCH) Study. Diabetologia 2011;54:87–92
11. Gillman MW, Rifas-Shiman S, Berkey CS,
Field AE, Colditz GA. Maternal gestational diabetes, birth weight, and adolescent obesity. Pediatrics 2003;111:e221–e226
12. Dabelea D, Hanson RL, Lindsay RS, et al. Intrauterine exposure to diabetes conveys risks
for type 2 diabetes and obesity: a study of discordant sibships. Diabetes 2000;49:2208–
2211
13. Boerschmann H, Pflüger M, Henneberger L,
Ziegler AG, Hummel S. Prevalence and predictors of overweight and insulin resistance in offspring of mothers with gestational diabetes
mellitus. Diabetes Care 2010;33:1845–1849
14. Tsadok MA, Friedlander Y, Paltiel O, Manor
O, Meiner V, Hochner H, et al. Obesity and blood
pressure in 17-year-old offspring of mothers
with gestational diabetes: insights from the
Jerusalem Perinatal Study. Exp Diabetes Res
2011;2011:906154
15. Lawlor DA, Fraser A, Lindsay RS, et al. Association of existing diabetes, gestational diabetes
and glycosuria in pregnancy with macrosomia
and offspring body mass index, waist and fat
mass in later childhood: findings from a prospective pregnancy cohort. Diabetologia 2010;
53:89–97
16. Baptiste-Roberts K, Nicholson WK, Wang
NY, Brancati FL. Gestational diabetes and subsequent growth patterns of offspring: the National Collaborative Perinatal Project. Matern
Child Health J 2012;16:125–132
17. Ehrlich SF, Rosas LG, Ferrara A, et al. Pregnancy glycemia in Mexican-American women
without diabetes or gestational diabetes and
programming for childhood obesity. Am J Epidemiol 2013;177:768–775
18. Kim SY, England JL, Sharma JA, Njoroge T.
Gestational diabetes mellitus and risk of childhood overweight and obesity in offspring: a systematic review. Exp Diabetes Res 2011;2011:
541308
19. Galic S, Oakhill JS, Steinberg GR. Adipose
tissue as an endocrine organ. Mol Cell Endocrinol 2010;316:129–139
20. Expert Panel on Detection, Evaluation, and
Treatment of High Blood Cholesterol in Adults.
Executive Summary of The Third Report of The
National Cholesterol Education Program (NCEP)
Expert Panel on Detection, Evaluation, And
Treatment of High Blood Cholesterol In Adults
3001
3002
Diabetes Care Volume 37, November 2014
Pregnancy Hyperglycemia and Daughter’s Obesity
(Adult Treatment Panel III). JAMA 2001;285:
2486–2497
21. Pacifico L, Anania C, Martino F, et al. Management of metabolic syndrome in children and
adolescents. Nutr Metab Cardiovasc Dis 2011;
21:455–466
22. Krieger N. Overcoming the absence of socioeconomic data in medical records: validation
and application of a census-based methodology. Am J Public Health 1992;82:703–710
23. Hiatt RA, Haslam SZ, Osuch J; Breast Cancer
and the Environment Research Centers. The
breast cancer and the environment research
centers: transdisciplinary research on the role
of the environment in breast cancer etiology.
Environ Health Perspect 2009;117:1814–1822
24. Deardorff J, Ekwaru JP, Kushi LH, et al. Father absence, body mass index, and pubertal
timing in girls: differential effects by family income and ethnicity. J Adolesc Health 2011;48:
441–447
25. Ferrara A, Weiss NS, Hedderson MM, et al.
Pregnancy plasma glucose levels exceeding the
American Diabetes Association thresholds, but
below the National Diabetes Data Group thresholds for gestational diabetes mellitus, are related to the risk of neonatal macrosomia,
hypoglycaemia and hyperbilirubinaemia. Diabetologia 2007;50:298–306
26. American Diabetes Association. Gestational
diabetes mellitus. Diabetes Care 2000;23(Suppl.
1):S77–S79
27. Kuczmarski R, Ogden CL, Guo S. 2000 CDC
Growth Charts for the United States: Methods
and Development. Atlanta, GA, Centers for Disease Control and Prevention, 2000
28. Marshall WA, Tanner JM. Variations in pattern of pubertal changes in girls. Arch Dis Child
1969;44:291–303
29. Nehring I, Chmitorz A, Reulen H, von Kries
R, Ensenauer R. Gestational diabetes predicts
the risk of childhood overweight and abdominal
circumference independent of maternal obesity. Diab Med 2013;30:1449–1456
30. Wright CS, Rifas-Shiman SL, Rich-Edwards
JW, Taveras EM, Gillman MW, Oken E. Intrauterine exposure to gestational diabetes, child adiposity, and blood pressure. Am J Hypertens
2009;22:215–220
31. Ritchie SA, Connell JM. The link between
abdominal obesity, metabolic syndrome and
cardiovascular disease. Nutr Metab Cardiovasc
Dis 2007;17:319–326
32. Pettitt DJ, Knowler WC, Bennett PH, Aleck
KA, Baird HR. Obesity in offspring of diabetic
Pima Indian women despite normal birth
weight. Diabetes Care 1987;10:76–80
33. Sullivan EL, Grove KL. Metabolic imprinting
in obesity. Forum Nutr 2010;63:186–194
34. Agha M, Agha RA, Sandell J. Interventions
to reduce and prevent obesity in pre-conceptual
and pregnant women: a systematic review and
meta-analysis. PLoS ONE 2014;9:e95132