Aleitamento materno e adiposidade adulta (JPed 2014) - Artigo apresentado em Reunião Científica da Liga Acadêmica de Pediatria da UFRN - LAPED UFRN - Natal - RN - Brasil.

1. Breastfeeding, Early Nutrition, and Adult Body Fat
Sandrine Peneau, PhD1,2,3,4,5,6
, Serge Hercberg, MD, PhD1,2,3,4,5,6,7,8,9
, and Marie-Franc¸oise Rolland-Cachera, PhD1,2,3,4,5,6
Objective To examine the association between breastfeeding and adult body fatness, adjusting for nutritional
intake in early childhood.
Study design Nutritional intakes of 73 healthy infants born in 1984 who participated in the 2-decade-long Lon-
gitudinal Study of Nutrition and Growth in Children (Etude Longitudinale Alimentation Nutrition Croissance des En-
fants [ELANCE]) were estimated at age 10 months and again at age 2 years. Breastfeeding was defined as any
breastfeeding, including partial breastfeeding, regardless of duration. At age 20 years, weight, height, subscapular
skinfold thickness (SF), and fat mass (assessed via bioelectrical impedance analysis) were measured.
Results In this sample, 64% of the children had been breastfed. In linear regression models adjusted for mother’s
body mass index and father’s profession, breastfeeding was not associated with any of the body fat measurements
at 20 years (all P .05). After adding nutritional intake variables (total energy and % energy from nutrients) to the
models, breastfeeding became significantly associated with lower SF at 20 years. In particular, breastfed subjects
had significantly lower % SF at 20 years after adjustment for energy and % fat intakes at 2 years of age,
(b = À28.25% SF; 95% CI, À50.28% to À6.21%; P = .013) or when adjusting for energy and % carbohydrates
at 2 years of age (b = À28.27% SF; 95% CI, À50.64% to À5.90%; P = .014).
Conclusion Breastfeeding was not associated with adult body fatness taking into account the usual confounding
factors. However, after also adjusting for nutritional intake covariates, a protective effect of breastfeeding emerged.
Early nutrition needs to be taken into account when examining the long-term health effects of breastfeeding. (J Pe-
diatr 2014;164:1363-8).
T
he role of breastfeeding in predicting later body fatness has generated substantial interest in recent years. Numerous
studies have shown that breastfeeding is protective against overweight and obesity,1-3
but others have not supported
this effect.4,5
Conflicting findings among studies might be related to the type of statistical adjustment,1,6,7
body fat in-
dicator of interest,8
or age at outcome assessment.7
Early nutrition is an important factor that may play a role in the association between breastfeeding and later body fatness. An
imbalanced diet in early life can alter growth and body fat development. In industrialized countries, infants’ diets are generally
characterized by high protein and low fat intake,9-12
and several studies have shown a positive correlation between protein
intake and body fat development.10,13
Recently, data reported from the Longitudinal Study of Children’s Nutrition and Growth
(Etude Longitudinale Alimentation Nutrition Croissance des Enfants [ELANCE]) showed an association between low fat intake
in early life and increased body fatness in adulthood.12
These data suggest that early fat restriction, like poor nutrition in other
contexts, could program later overweight.14
A potential influence of nutritional factors, such as maternal feeding restriction of the child15
and age at introduction of spoon
feeding,16
on the association between breastfeeding and later body fat has been suggested previously. However, to our knowledge,
no study to date has investigated the specific contribution of macronutrient intake during infancy. In addition, most previous
studies investigated the impact of breastfeeding on later overweight or obesity risk,3,17
while anthropometric characteristics,
such as body mass index (BMI), skinfold thickness (SF), and body composition measurements, have been poorly investigated.
The present analysis is based on the ELANCE cohort of healthy French children followed from early life through age 20 years. We
investigated whether breastfeeding is correlated with body fatness in adulthood, taking into account early nutritional factors.
Methods
Healthy infants and toddlers born in 1984 were invited for a free health exami-
nation at age 10 months, 2 years, and 4 years at a health center for children
(health center study). Subjects who had completed at least 2 such visits (at 10
From the 1
Research Team Nutritional Epidemiology,
Center of Research in Epidemiology and Biostatistics,
University of Paris 13, Sorbonne-Paris, Paris, France;
2
Institut National de la Sante et de la Recherche
Medicale, U1153, Bobigny, France; 3
National Institute of
Health and Medical Research; 4
National Conservatory of
Arts and Crafts, Bobigny, France; 5
University of Paris 5;
6
University of Paris 7; 7
University of Paris 13, Sorbonne-
Paris, Surveillance Unit and Nutritional Epidemiology;
8
Institute for Public Health Surveillance; 9
Department of
Public Health, Avicenna Hospital, Bobigny, France
Supported by Benjamin Delessert Institute (to perform
the data collection at age 20 years). The authors declare
no conflicts of interest.
0022-3476/$ - see front matter. Copyright ª 2014 Elsevier Inc.
All rights reserved.
http://dx.doi.org/10.1016/j.jpeds.2014.02.020
BIA Bioelectrical impedance analysis
BMI Body mass index
ELANCE Etude Longitudinale Alimentation Nutrition Croissance des Enfants (Longitudinal Study
of Nutrition and Growth in Children)
FM Fat mass
SF Skinfold thickness
1363

2. months and 2 years or at 10 months and 4 years; n = 222)
were invited to participate in the ELANCE prospective study
on nutrition and growth.13
A total of 126 mothers consented
to participate in the follow-up, which included regular home
visits by a dietician. Thus, 126 children were examined at age
6 years, 112 were examined at 8 years, 104 were examined at
10 years, 97 were examined at 14 years, and 94 were examined
at 16 years. At age 20 years, the cohort was invited to partic-
ipate in a health examination at a health center for adults, and
73 subjects completed this visit. All of the 73 subjects exam-
ined at age 20 years had been examined at age 10 months, and
68 had been examined at age 2 years.
The Ethics Committee of the Paris Cochin Hospital
approved this study (CCPPRB 2179). Written informed con-
sent was obtained from the parents during the follow-up and
from each participant at age 20 years.
Detailed information on breastfeeding (exclusive or par-
tial; duration) was obtained from the mothers via face-to-
face interviews when the child was 10 months of age. In the
present analysis, breastfeeding was defined as any kind of
breastfeeding, including partial breastfeeding regardless of
duration. Breastfeeding was never initiated in the non–
breastfed group. Food intake was assessed by a dietitian via
the dietary history method, as described previously.13
In
brief, the mother (occasionally the father) was interviewed
for approximately 45 minutes about the child’s eating pat-
terns during the month before the interview. For a child
fed outside the home, information about the child’s intake
was obtained from the caregiver.
Body measurements at age 20 years were performed in the
healthcenter foradults bya trained investigatorfollowing stan-
dard procedures.18
All measurements were obtained in the
morningwiththesubjectinafastingstateandwearinglightun-
derwear. Heightwas measured tothe nearest 0.1 cm witha Seca
240 wall-mounted stadiometer (Seca, Hamburg, Germany).
Weight was measured as part of the bioelectrical impedance
analysis(BIA)proceduretothenearest0.1kg.SFwasmeasured
at the subscapular site to the nearest 0.1 mm with a Harpenden
calliper (British Indicators Ltd, West Sussex, United
Kingdom). Body composition was assessed with an 8-
electrode Tanita BC-418 segmental body composition analyzer
(Tanita, Tokyo, Japan) after a 30-minute rest. Fat mass (FM)
was derived from the analyzer manufacturer’s equations.
Several relevant confounders of the association between
breastfeeding and subsequent body fat are identified in the
literature. In the present analysis, we considered mother’s
BMI and father’s profession (unskilled/semiskilled vs
skilled/professional). Mother’s weight and height and father’s
profession were self-declared during the interview at age 10
months. Given recent findings showing that early nutrition
can affect later body fatness,12
nutritional intake at ages 10
months and 2 years (ie, total energy and % energy from
each nutrient) were also added into the models.
Statistical Analyses
BMI was calculated (weight in kilograms divided by the
squared height in meters) for each subject. Comparisons
of characteristics between subjects who were followed up
to age 20 years and those who were lost to follow-up and
between the sexes were based on the Student t test or c2
test, as appropriate. The Kolmogorov-Smirnov test was
used to assess the skewness of the variables. Only the sub-
scapular SF distribution was skewed; the Wilcoxon Mann-
Whitney test was used for this variable. Before the fitting
of linear regression models, the SF was log(e)-transformed
and multiplied by 100. Thus, the regression coefficients
are in units of % SF per unit of each independent vari-
able.19
Comparisons of nutritional intake at age 10 months
and 2 years according to the presence or absence of breast-
feeding were based on the Student t test or c2
test, as
appropriate.
The associations between breastfeeding (yes vs no) and
adult body measurements (BMI, SF, and FM [BIA]) were
analyzed using linear regression models. Regression
models were performed with various adjustments: sex
only (model 1), sex + mother’s BMI + father’s profession
(model 2), and sex + mother’s BMI + father’s profession +
energy intake at age 10 months (model 3) or at 2 years of
age (model 5). Models 3 and 5 were further adjusted for
protein (models 4a and 6a), lipid (models 4b and 6b), or
carbohydrate (models 4c and 6c) intakes, expressed as %
energy. In all models, total energy intake was maintained
in accordance with the adjusted multivariate nutrient den-
sity model proposed by Willett et al.20
In the analysis with
FM as the outcome variable, height at age 20 years was also
added to the models. The strength and direction of the as-
sociations between nutrient intake at age 10 months and 2
years and adult body measurements were also considered
in these different models.
All outcome variables were available for all subjects, and no
variable used for adjustments was missing. All analyses were
performed using SPSS 12.0.1 (SPSS Inc, Chicago, Illinois),
with the significance level set at P .05 (2-sided).
Results
Subjects who were followed up to age 20 years (n = 73) and
those who were lost to follow-up had similar characteristics
in early life. Weight and length at birth; nutritional intake
(energy and macronutrients) at ages 10 months and 2 years,
and breastfeeding (frequency and duration) did not differ be-
tween the 2 groups (all P .05). The 2 groups also had similar
BMI and SF at all ages during follow-up. Characteristics of
the 73 subjets who participated in the follow-up from age
10 months to age 20 years are presented in Table I. Two-
thirds of the children had been breastfed, with a maximum
duration of breastfeeding of 7.5 months. Nutritional
intakes at ages 10 months and 2 years have been published
previously.12
In the initial sample of 222 children examined in the health
centers (health center study), nutritional intake at age 10
months did not differ between breastfed and non–breastfed
children, but at age 2 years, fat intake (% energy) was signif-
icantly lower in the breastfed children (% fat: 33.1% Æ 5.3%
THE JOURNAL OF PEDIATRICS www.jpeds.com Vol. 164, No. 6
1364 Peneau, Hercberg, and Rolland-Cachera

3. vs 34.6% Æ 4.5%; P = .048), and carbohydrate intake (% en-
ergy) tended to be higher in the breastfed children (50.5% Æ
6.4% vs 48.8% Æ 5.1%; P = .051). No between-group differ-
ences in protein intake were observed.
The same trends were observed in the 73 children of the
follow-up sample. At age 2 years, fat intake (% energy) was
31.9 Æ 5.6 in breastfed and 33.5 Æ 3.9 in non-breastfed
(P = .23) children, while carbohydrate intake (% energy)
was 51.6 Æ 6.7 in breastfed and 49.7 Æ 3.8 in non-breastfed
(P = .22) children. The absence of significant associations
in the follow-up sample can be explained by the loss of power
owing to the reduced sample size.
The associations between breastfeeding and later BMI, FM
(BIA), and SF are presented Table II. The associations
between being breastfed and later body fatness varied
according to the confounding factors taken into account and
the outcome measure considered. The associations between
breastfeeding and BMI or FM (BIA) were not significant
(P = .63 and .33, respectively) (Table II). However,
breastfeeding tended to be negatively associated with FM
after adjusting for both energy and lipid intake (P = .066) or
for both energy and carbohydrate intake (P = .079).
Breastfeeding tended to be negatively associated with SF (P
= .055), and the association became significant after adjusting
for nutritional intake at 2 years. A protective effect of breast-
feeding appeared when both energy and lipids (P = .013) or
both energy and carbohydrates (P = .014) were added into
the models (Table II).
The strengths and directions of the associations among the
various nutrients at 2 years and adult SF are as follows. In
model 6a, protein intake was not associated with SF (b =
0.47% SF; 95% CI, 0.02%-0.21%; P = .84), in model 6b,
higher lipid intake (%) was significantly associated with
lower %SF (b = À2.30% SF; 95% CI, À4.41% to À0.18%;
P = .034) and in model 6c, higher intake of carbohydrates
(%) tended to be associated with higher %SF (b = 1.67%
Table I. Sex-specific characteristics in infancy and adulthood of the subjects of the French ELANCE longitudinal study
(n = 73)
All (n = 73) Males (n = 40) Females (n = 33) P*
Measurements in infancy
Breastfeeding
Prevalence of breastfeeding, % 64 72.5 54.5 .089
Duration of breastfeeding, mo, mean (SD)
Full sample 1.6 (1.9) 1.9 (1.9) 1.3 (1.9) .19
Breastfed only 2.47 (1.83) 2.53 (1.76) 2.38 (1.99) .79
Father’s occupation at child’s birth, %
Unskilled/semiskilled 61 57.5 63.6 .64
Skilled 39 42.5 36.4
Mother’s BMI, kg/m2
, mean (SD) 21.4 (2.7) 21.4 (2.8) 21.3 (2.5) .90
Measurements at age 20 years
Anthropometry
Weight, kg, mean (SD) 64.9 (13.8) 71.9 (12.1) 56.4 (10.6) .001
Height, m, mean (SD) 1.72 (0.10) 1.79 (0.06) 1.63 (0.07) .001
BMI, kg/m2
, mean (SD) 21.9 (3.5) 22.5 (3.7) 21.2 (3.1) .103
Subscapular SF, % 15.9 (8.2) 14.4 (7.8) 17.6 (8.5) .035†
BIA
FM, kg, mean (SD) 11.2 (6.7) 9.3 (5.9) 13.5 (7) .007
*P values showing differences between male and female subjects, based on the Student t test or c2
test, as appropriate.
†P value based on the Wilcoxon Mann-Whitney test.
Table II. Multiple linear regression models for breastfeeding predicting anthropometry and body composition at age 20
years, adjusted for the usual confounding factors and nutritional intakes at age 10 months (n = 73) or age 2 years (n = 68)
in the ELANCE longitudinal study
Breastfeeding (yes vs no) BMI, kg/m2
Subscapular SF, %* FM (BIA), kg†
Model Adjustments b (95% CI) P b (95% CI) P b (95% CI) P
1 Sex À0.029 (À1.73 to 1.67) .97 À19.18 (À42.5 to 4.14) .11 À0.64 (À3.81 to 2.53) .69
2 Model 1 + mothers’ BMI + father’s occupation À0.413 (À2.12 to 1.29) .63 À22.58 (À45.6 to 0.47) .055 À1.52 (À4.64 to 1.60) .33
Adjustment for nutritional intake at age 10 months
3 Model 2 + energy, kcal À0.431 (À2.11 to 1.25) .61 À22.70 (À45.8 to 0.40) .054 À1.54 (À4.63 to 1.56) .32
4a Model 3 + proteins, % À0.228 (À1.95 to 1.49) .79 À20.16 (À43.9 to 3.57) .094 À1.14 (À4.33 to 2.04) .48
4b Model 3 + lipids, % À0.606 (À2.26 to 1.05) .47 À23.33 (À46.7 to 0.06) .051 À1.89 (À4.94 to 1.16) .22
4c Model 3 + carbohydrates, % À0.618 (À2.32 to 1.09) .47 À22.74 (À46.4 to 0.98) .060 À1.92 (À5.07 to 1.22) .23
Adjustment for nutritional intake at age 2 years
5 Model 2 + energy, kcal À0.772 (À2.39 to 0.85) .34 À25.35 (À47.88 to À2.83) .028 À2.36 (À5.44 to 0.72) .13
6a Model 5 + proteins, % À0.721 (À2.36 to 0.92) .38 À25.12 (À47.95 to À2.30) .032 À2.25 (À5.36 to 0.86) .15
6b Model 5 + lipids, % À0.891 (À2.52 to 0.74) .28 À28.25 (À50.28 to À6.21) .013 À2.83 (À5.86 to 0.20) .066
6c Model 5 + carbohydrates, % À0.865 (À2.51 to 0.78) .30 À28.27 (À50.64 to À5.90) .014 À2.76 (À5.86 to 0.33) .079
*Percent subscapular SF change between non–breastfed infants and breastfed infants.23
†For FM (BIA), height at age 20 years was added into all models.
June 2014 ORIGINAL ARTICLES
Breastfeeding, Early Nutrition, and Adult Body Fat 1365

4. SF; 95% CI, À0.20% to 3.54%; P = .079). These results sug-
gest that lipid intake had the strongest influence on the asso-
ciation between breastfeeding and SF.
Discussion
In the present study, there was no significant association be-
tween breastfeeding and BMI. Although adjustment for early
nutrition improved the strength of the association, it re-
mained nonsignificant. These results are consistent with
other studies reporting a relatively limited long-term impact
of breastfeeding.7
Here we found a nearly significant associ-
ation between breastfeeding and SF, taking into account
the mother’s BMI and the father’s occupation. Only a few
previous studies in this area have focused on SF. In line
with our data, one study suggested a trend toward lower ad-
ipose tissue values measured by SF in adolescent girls who
had been breastfed for more than 3 months compared with
those who had been breastfed for less than 3 months or not
at all.21
Other studies have indicated no association between
breastfeeding and SF measurements in adults, however.22,23
Finally, our data show no significant association between
breastfeeding and FM (BIA), consistent with previous re-
ports.23
Other relevant studies using dual-energy X-ray ab-
sorptiometry21,24
or densitometry22
reported limited
protective effects of breastfeeding. These findings indicate
that the impact of breastfeeding can vary according to the
outcome measure used.
Several factors are associated with obesity. Socioeconomic
status, parental BMI, and smoking are the most important
confounders taken into account in studies on the long-
term effects of breastfeeding.1,3,7,17
In addition to these fac-
tors, early nutrition also may affect later body fat. In the
present study, we examined the association between breast-
feeding and adult body fat measurements taking into account
nutritional intake at ages 10 months and 2 years as well as
mother’s BMI and father’s profession. Our results reveal sig-
nificant negative associations between breastfeeding and
body fatness when intakes of energy and macronutrients
(% energy from lipids or carbohydrates) at age 2 years were
added into our models. The associations became significant
for SF, and a tendency was observed for FM (BIA). In these
multivariate models, fat intake was the only confounding
macronutrient significantly associated with adult SF. The as-
sociation was negative between fat intake at age 2 years and
adult SF, and only a positive trend was observed between car-
bohydrates and SF. This positive trend between carbohy-
drates and SF might stem from countering low fat intakes,
consistent with the decline in dietary lipid content recorded
over several decades.25,26
The foregoing results demonstrate the importance of tak-
ing into account macronutrient intake, particularly lipids,
when investigating the association between breastfeeding
and body fat development. Several previous studies have at-
tempted to take into account the influence of early-life nutri-
tion on the association between breastfeeding and later body
fatness15,16,27
; however, to the best of our knowledge, no
study has quantitatively investigated macronutrient intake
during infancy.
We previously showed that early dietary fat restrictions
could be a risk factor for later overweight and leptin resis-
tance.12
These results suggested that the consumption of a
low energy-dense diet in early life could have promoted
metabolic adaptations to prevent underweight. Such an
anticipatory strategy could subsequently increase the suscep-
tibility to overweight, however.14
It has been consistently shown that a low-fat diet during
childhood is associated with greater BMI,28
and a recent re-
view concluded that a relatively high-fat diet is not harmful
to health later in life.29
These results are consistent with ani-
mal studies showing an association between early low fat
intake and increased body fatness.30
Because breastfeeding
is associated with lower SF, and because low fat intake is asso-
ciated with higher SF, then a low-fat diet after breastfeeding
may have a deleterious impact, countering the benefit of
breastfeeding.
Although advice for breastfeeding is common, informa-
tion on fat intake during the weaning period is not clear. A
study of mothers’ attitudes toward healthy eating regarding
their infants reveals some misconceptions, with 87% of
mothers believing that a low-fat intake is very important,31
with low-fat milk the most common type of milk given to in-
fants aged 1 year11,12
and 2 years.12
The most frequently cited
reasons for mothers feeding low-fat milk were the perception
that low-fat milk has less fat than cow’s (whole) milk and a
physician’s recommendation or suggestion.32
Considering
our initial sample of children examined in the health centers,
it appeared that mothers who breastfed provided diets with
less dietary fat at age 2 years compared with mothers who
did not breastfeed, perhaps because they were more health
conscious1
and believed that restricting fat is healthier.31
Consequently, breastfeeding could be followed by inade-
quate nutrient intake. Indeed, the high fat content of human
milk is adapted to the nutritional needs of the young child,
and % fat intake should be decreased gradually, from 40%-
60% at 6 months to 35% at 2 years.33
However, in our study
cohort, fat intake dropped sharply after weaning and subse-
quently increased (from 28% at age 10 months to 32% at
age 2 years),12
as has been reported in previous studies
from many other countries.9-11,34
A limitation of the present study is its small sample size,
which could have led to insufficient statistical power for de-
tecting existing associations. In particular, despite the large
decrease in adult FM associated with breastfeeding, the asso-
ciation did not reach statistical significance. In addition, a
significant number of subjects were lost to follow-up. Thus,
our results might not be generalizable to all types of popula-
tions and need to be replicated in different contexts. More-
over, several relevant confounders of the association
between breastfeeding and subsequent body fat, such as
gestational weight gain and smoking habits, which could
interfere with the results, were missing in our study. Breast-
feeding was defined as any kind of breastfeeding, including
partial breastfeeding, regardless of duration. The broad
THE JOURNAL OF PEDIATRICS www.jpeds.com Vol. 164, No. 6
1366 Peneau, Hercberg, and Rolland-Cachera

5. definition of breastfeeding used in this study and the rela-
tively short duration of breastfeeding (3 months on
average) might have influenced the results; however, the
prevalence and duration of breastfeeding seen in our cohort
were close to previously reported French national esti-
mates.35
In addition, breastfeeding was assessed retrospec-
tively at age 10 months, which might have led to potential
misclassification.36
The weak associations with nutritional
intake seen at age 10 months may be explained by the highly
homogeneous composition of the diet, which is often based
on infant formula in the first months of life. Outcome mea-
surements were recorded in a young adult population with a
relatively low mean BMI. However, as stated previously, BMI
does not accurately represent body composition.37
In turn,
the BIA method has some limitations compared with dual-
energy X-ray absorptiometry, including a tendency to under-
estimate FM.38
An imbalanced diet consumed after the breastfeeding
period could counteract the beneficial effects of breastfeed-
ing. This effect may account for the conflicting results on
the role of breastfeeding reported in the literature. In addi-
tion, the association between breastfeeding and later body
fatness clearly depends on the body measurement used as
the outcome. Previously published results investigating the
association between breastfeeding and body fat should be re-
considered in view of the important yet often lacking role of
nutrition in the first years of life. n
We are indebted to the participants and their parents and to the field
and laboratory staff for their involvement and contribution. We thank
Michele Deheeger, who recorded the data throughout the whole
ELANCE study.
Submitted for publication Jul 26, 2013; last revision received Jan 17, 2014;
accepted Feb 6, 2014.
References
1. Horta BL, Bahl R, Martines JC, Victora CG. Evidence on the long-term
effects of breastfeeding: systematic reviews and meta-analysis. Geneva,
Switzerland: World Health Organization. Available from: http://
whqlibdoc.who.int/publications/2007/9789241595230_eng.pdf. Ac-
cessed January 14, 2014.
2. Monasta L, Batty GD, Cattaneo A, Lutje V, Ronfani L, van Lenthe FJ,
et al. Early-life determinants of overweight and obesity: a review of sys-
tematic reviews. Obes Rev 2010;11:695-708.
3. Owen CG, Martin RM, Whincup PH, Smith GD, Cook DG. Effect of in-
fant feeding on the risk of obesity across the life course: a quantitative
review of published evidence. Pediatrics 2005;115:1367-77.
4. Casazza K, Fontaine KR, Astrup A, Birch LL, Brown AW, Bohan
Brown MM, et al. Myths, presumptions, and facts about obesity. N
Engl J Med 2013;368:446-54.
5. Cope MB, Allison DB. Critical review of the World Health Organiza-
tion’s (WHO) 2007 report on “evidence of the long-term effects of
breastfeeding: systematic reviews and meta-analysis” with respect to
obesity. Obes Rev 2008;9:594-605.
6. Clifford TJ. Breastfeeding and obesity. BMJ 2003;327:879-80.
7. Owen CG, Martin RM, Whincup PH, Davey-Smith G, Gillman MW,
Cook DG. The effect of breastfeeding on mean body mass index
throughout life: a quantitative review of published and unpublished
observational evidence. Am J Clin Nutr 2005;82:1298-307.
8. Wells JC, Chomtho S, Fewtrell MS. Programming of body composition
by early growth and nutrition. Proc Nutr Soc 2007;66:423-34.
9. Butte NF, Fox MK, Briefel RR, Siega-Riz AM, Dwyer JT, Deming DM,
et al. Nutrient intakes of US infants, toddlers, and preschoolers meet
or exceed dietary reference intakes. J Am Diet Assoc 2010;110:S27-37.
10. Hoppe C, Molgaard C, Thomsen BL, Juul A, Michaelsen KF. Protein
intake at 9 mo of age is associated with body size but not with body
fat in 10-y-old Danish children. Am J Clin Nutr 2004;79:494-501.
11. Kyttala P, Erkkola M, Kronberg-Kippila C, Tapanainen H, Veijola R,
Simell O, et al. Food consumption and nutrient intake in Finnish 1- to
6-year-old children. Public Health Nutr 2010;13:947-56.
12. Rolland-Cachera MF, Maillot M, Deheeger M, Souberbielle JC,
Peneau S, Hercberg S. Association of nutrition in early life with body
fat and serum leptin at adult age. Int J Obes (Lond) 2013;37:1116-22.
13. Rolland-Cachera MF, Deheeger M, Akrout M, Bellisle F. Influence of
macronutrients on adiposity development: a follow-up study of nutri-
tion and growth from 10 months to 8 years of age. Int J Obes Relat Metab
Disord 1995;19:573-8.
14. Hales CN, Barker DJ. The thrifty phenotype hypothesis. Br Med Bull
2001;60:5-20.
15. Taveras EM, Rifas-Shiman SL, Scanlon KS, Grummer-Strawn LM,
Sherry B, Gillman MW. To what extent is the protective effect of breast-
feeding on future overweight explained by decreased maternal feeding
restriction? Pediatrics 2006;118:2341-8.
16. Schack-Nielsen L, Sorensen TI, Mortensen EL, Michaelsen KF. Late intro-
duction of complementary feeding, rather than duration of breastfeeding,
may protect against adult overweight. Am J Clin Nutr 2010;91:619-27.
17. Arenz S, von Kries R. Protective effect of breastfeeding against obesity in
childhood: can a meta-analysis of observational studies help to validate
the hypothesis? Adv Exp Med Biol 2005;569:40-8.
18. Lohman TG, Roche AF, Martorell R, eds. Anthropometric standardiza-
tion. reference manual. Champaign (IL): Human Kinetics Books; 1988.
19. Cole TJ. Sympercents: symmetric percentage differences on the 100
log(e) scale simplify the presentation of log transformed data. Stat
Med 2000;19:3109-25.
20. Willett WC, Howe GR, Kushi LH. Adjustment for total energy intake in
epidemiologic studies. Am J Clin Nutr 1997;65:1220S-8S.
21. TulldahlJ,PetterssonK,Andersson SW,HulthenL.Modeofinfant feeding
and achieved growth in adolescence: early feeding patterns in relation to
growth and body composition in adolescence. Obes Res 1999;7:431-7.
22. Pirila S, Saarinen-Pihkala UM, Viljakainen H, Turanlahti M,
Kajosaari M, Makitie O, et al. Breastfeeding and determinants of adult
body composition: a prospective study from birth to young adulthood.
Horm Res Paediatr 2012;77:281-90.
23. Victora CG, Barros F, Lima RC, Horta BL, Wells J. Anthropometry and
body composition of 18 year old men according to duration of breast-
feeding: birth cohort study from Brazil. BMJ 2003;327:901.
24. Toschke AM, Martin RM, von Kreis R, Wells J, Smith GD, Ness AR. In-
fant feeding method and obesity: body mass index and dual-energy
X-ray absorptiometry measurements at 9-10 y of age from the Avon Lon-
gitudinal Study of Parents and Children (ALSPAC). Am J Clin Nutr
2007;85:1578-85.
25. Alexy U, Sichert-Hellert W, Kersting M. Fifteen-year time trends in en-
ergy and macronutrient intake in German children and adolescents: re-
sults of the DONALD study. Br J Nutr 2002;87:595-604.
26. Gibson S. Trends in energy and sugar intakes and body mass index be-
tween 1983 and 1997 among children in Great Britain. J Hum Nutr
Diet 2010;23:371-81.
27. Hediger ML, Overpeck MD, Kuczmarski RJ, Ruan WJ. Association be-
tween infant breastfeeding and overweight in young children. JAMA
2001;285:2453-60.
28. Alexy U, Sichert-Hellert W, Kersting M, Schultze-Pawlitschko V. Pattern
of long-term fat intake and BMI during childhood and adolescence: re-
sults of the DONALD Study. Int J Obes Relat Metab Disord 2004;28:
1203-9.
29. Agostoni C, Caroli M. Role of fats in the first two years of life as related
to later development of NCDs. Nutr Metab Cardiovasc Dis 2012;22:
775-80.
June 2014 ORIGINAL ARTICLES
Breastfeeding, Early Nutrition, and Adult Body Fat 1367

6. 30. Shahkhalili Y, Mace K, Moulin J, Zbinden I, Acheson KJ. The fat:carbo-
hydrate energy ratio of the weaning diet programs later susceptibility to
obesity in male Sprague Dawley rats. J Nutr 2011;141:81-6.
31. Morgan JB, Kimber AC, Redfern AM, Stordy BJ. Healthy eating for in-
fants: mothers’ attitudes. Acta Paediatr 1995;84:512-5.
32. Ryan AS, Martinez GA, Krieger FW. Feeding low-fat milk during in-
fancy. Am J Phys Anthropol 1987;73:539-48.
33. Food and Agriculture Organization of the United Nations. Fats and fatty
acids in human nutrition: report of an expert consultation. FAO food
and nutrition paper 91. Rome, Italy: Food and Agriculture Organization
of the United Nations; 2010. p. 1-166.
34. Simell O, Niinikoski H, Ronnemaa T, Lapinleimu H, Routi T,
Lagstrom H, et al. Special Turku Coronary Risk Factor Intervention
Project for Babies (STRIP). Am J Clin Nutr 2000;72:1316S-31S.
35. Crost M, Kaminski M. L’allaitement maternel a la maternite en
France en 1995: enqu^ete nationale perinatale. Arch Pediatr 1998;5:
1316-26.
36. Huttly SR, Barros FC, Victora CG, Beria JU, Vaughan JP. Do mothers
overestimate breastfeeding duration? An example of recall bias from a
study in southern Brazil. Am J Epidemiol 1990;132:572-5.
37. Okorodudu DO, Jumean MF, Montori VM, Romero-Corral A,
Somers VK, Erwin PJ, et al. Diagnostic performance of body mass index
to identify obesity as defined by body adiposity: a systematic review and
meta-analysis. Int J Obes (Lond) 2010;34:791-9.
38. Volgyi E, Tylavsky FA, Lyytikainen A, Suominen H, Alen M, Cheng S.
Assessing body composition with DXA and bioimpedance: effects of
obesity, physical activity, and age. Obesity (Silver Spring) 2008;16:
700-5.
THE JOURNAL OF PEDIATRICS www.jpeds.com Vol. 164, No. 6
1368 Peneau, Hercberg, and Rolland-Cachera