During the last decade there has been increasing interest in whether environmental chemical exposures may contribute to the rising prevalence of obesity. Experimental evidence suggests that exposure to up to 20 different chemical classes, including bisphenol A (BPA), during the critical periods of development in fetal and early life may increase the risk for weight gain later in childhood or adulthood.1 However, there are few epidemiologic data on early-life chemical exposures that may be obesogenic. Such studies have focused mainly on the effects of nicotine2 and, to a lesser extent, of persistent organic pollutants (eg, dichlorodiphenyl-dichloroethylene [DDE] and polychlorinated biphenyls)3 on weight gain.
BPA is a chemical used extensively worldwide in the manufacture of plastic polymers such as polycarbonate plastics and epoxy resins, which are found in many consumer products (eg, toys, food and beverage containers, water supply pipes, medical tubing,4 and cigarette filters5). BPA has been detected at measurable concentrations in urine samples in almost all persons tested around the world.4–7 The main route for human exposure is considered to be dietary ingestion.4 Detectable BPA levels have also been measured in placental and amniotic fluids and human breast milk,4 suggesting that exposure starts in utero and may continue after birth via breastfeeding practices.
Evidence from studies in rodents suggests that developmental exposure to BPA at environmentally relevant doses may alter early adipogenesis and increase fat mass later in life.8–10 BPA effects on weight homeostasis have been further suggested to be sex-specific,9,10 and dose-dependent with stronger effects at lower doses.11 Despite the accumulating evidence from experimental studies, few human studies have evaluated the effects of early-life BPA exposure on obesity,12,13 of which only one included a prospective analysis.13 The two previous cross-sectional analyses both showed increasing body mass index (BMI) with increasing BPA levels in children and adolescents,12,13 whereas the prospective analyses observed a reduction of BMI in girls from 2 to 9 years of age with increasing prenatal BPA exposure levels.13
There is substantial evidence that obesity risk may begin very early in life; rapid weight gain in the first months after birth has been consistently associated with a subsequent increased risk of obesity in childhood14 and later in life.15,16 However, the effect of prenatal BPA exposure on rapid weight gain in the first months of life has not been evaluated in prospective studies. In this study, we therefore examined the effects of prenatal BPA exposure on rapid growth in the first 6 months of life and on other obesity-related anthropometric measurements (ie, waist circumference and BMI) later in infancy and early childhood.
The population-based birth cohort study INMA (“INfancia y Medio Ambiente”—Environment and Childhood)17 recruited 657 pregnant women in the Spanish region of Sabadell in 2004 to 2006. All women were enrolled in the 1st trimester of pregnancy at the primary healthcare center. The inclusion criteria were age at least 16 years, intention to give birth in the reference hospital, no problems in communication, singleton pregnancy, and no assisted conception.17 Maternal urine samples were collected in the 1st trimester (around week 12) and the 3rd trimester (around week 32) of pregnancy. Of 657 mother–infant pairs initially enrolled, 619 participated in the follow-up conducted at the time of delivery. For 469 children, BPA concentrations were measured in maternal urine samples in both the 1st and 3rd trimesters of pregnancy; of those, 424 had available information on both BPA concentrations in the two trimesters and child growth data. Of these, we included in the main analysis 402 children born at term (≥37 weeks of gestation) and with complete data on key covariates. Children born preterm are known to follow different catch-up growth patterns than children born at term,18 and the growth curves we have applied (see below) may not be valid for these children; thus, we excluded preterm births from the main analysis (n = 20).
Information was collected by questionnaires in the 1st and 3rd trimesters of pregnancy, at delivery, and when the child was 6 months, 14 months, and 4 years old. Information was obtained on parental sociodemographic characteristics (age, education, social class coded based on occupation according to the International Standard Classification of Occupatons-88 system),19 maternal and paternal weight and height (reported by the mothers), and other maternal characteristics (including medical history, parity, medication, alcohol consumption in pregnancy, and active or passive smoking). Maternal diet was assessed in the 1st and 3rd trimesters using a 101-item food frequency questionnaire validated for use in Spanish adults.20 Child feeding practices and sedentary activities were reported in postnatal questionnaires. We defined exclusive breastfeeding duration as the number of weeks during which children received breast milk with or without supplementary nonmilk liquids such as water, and with no formula milk, and no solid foods.21
Informed consent was signed by all mothers. This study was approved by the ethics committee of the Hospital del Mar Research Institute and conducted according to principles of the Declaration of Helsinki.22
Repeated weight measures from birth to 6 months of age were extracted from medical records. For children without weight measurement available within ±14 days of their exact 6-month anniversary (n = 67, 17% of the main analysis sample), we used preexisting sex-specific growth models developed for early infancy to predict weight at 6 months of age.23 We compared six growth models: the Count, Kouchi, 1st- and 2nd-order Reed, Jenss, and I-component of Karlberg models.23 The best fit was obtained for the 2nd-order Reed model for both girls and boys (results not shown). Age- and sex-specific Z scores for weight at birth and at 6 months of age were calculated using the World Health Organization (WHO) referent.24 Rapid growth from birth to 6 months of age was defined as a Z score weight gain greater than 0.67 standard deviation (SD).16 Children with a Z score weight gain equal to or below 0.67 SD were characterized as slow/average growers.
Child weight (nearest gram) and length (nearest 0.1 cm), at 14 months and 4 years of age, were measured by trained staff using standard protocols (without shoes and in light clothing). Child BMI (weight/length)2 was used to estimate age- and sex-specific Z scores based on the WHO referent.24 Overweight was defined as a BMI Z score equal to or above the 85th percentile.24
At 14 months and 4 years of age, waist circumference was measured at the nearest 0.1 cm using an inelastic tape (SECA model 201; SECA, Hamburg, Germany). At 14 months, waist circumference was measured at the umbilicum while infants were in a supine position on a hard surface and at the time of minimal respiration. At 4 years, waist circumference was measured at the midpoint between the lowest rib margin and the iliac crest, in a standing position and after a gentle expiration. In the absence of a widely used validated reference for standardizing waist circumference measurements in very young children, we calculated sex- and age-specific Z scores using the mean of the population under study (waist circumference SD = [child waist circumference − sex- and age-specific population waist circumference mean]/sex- and age-specific population waist circumference SD). For this purpose child age was grouped into months.
For sensitivity analyses, we compared results defining overweight using weight-for-length age- and sex-specific Z scores instead of BMI Z scores24 as this metric is also often used in infancy. Analyses using the waist-to-height ratio (ie, waist circumference in cm/height in cm) versus the waist circumference Z scores adjusting models for height were also compared, as the former has recently been advocated as a marker of central obesity.25
Prenatal BPA Exposure
We measured BPA and creatinine concentrations in the two maternal spot-urine samples collected in 1st and 3rd trimesters of gestation. Urine samples were stored in polypropylene tubes at −20°C before analysis. Analyses were performed at the Department of Analytical Chemistry laboratory in the University of Cordoba (Spain). Total BPA (ie, free plus conjugated) was quantified using liquid chromatography mass spectrometry following methods described in detail elsewhere.26 The detection limit was 0.1 μg/L. All mothers included in the analysis population had detectable BPA concentrations except for two mothers with values below the limit of detection in the 3rd trimester of pregnancy. We replaced these values with half of the detection limit (ie, 0.1/2 = 0.05 μg/L). Creatinine was determined at the Echevarne laboratory of Barcelona (Spain) by the Jaffé method (kinetic with target measurement, compensated method) with Beckman Coulter reactive in AU5400 (IZASA®, Barcelona, Spain ).
BPA concentrations were divided by urinary creatinine levels to control for urine dilution (ie, creatinine-adjusted concentrations, hereafter). All creatinine-adjusted BPA concentrations were log10 transformed to obtain normal distributions, as the original distributions were right skewed. Single spot-urine concentrations may be appropriate only for measurement of very recent exposure to BPA because of the presumed rapid clearance of BPA from the body (half-life approximately 6 hours).5,27 As our main exposure variable we therefore used the average of the creatinine-adjusted BPA concentrations measured in the 1st and 3rd trimester, rather than the single-spot concentrations, as the best approximation to average BPA exposure throughout pregnancy. The separate single-spot measurements were analyzed as exposure variables in sensitivity analyses.
Linear regression models were used to estimate the β coefficients for the association between BPA concentrations and continuous obesity-related outcomes (weight change, waist circumference, and BMI Z scores). Generalized linear models were used to estimate the relative risks (RRs) for the association between BPA concentrations and binary outcomes (rapid growth and overweight). Generalized additive models (GAMs) were used to explore the shape of the relationships between BPA concentrations and other continuous covariates in the models (maternal age at delivery, parental BMI, maternal weight gain during pregnancy, and child age and height) and all outcome variables. GAMs indicated that the associations between BPA and the outcomes did not deviate from linearity (P-gain >0.10). Thus, we present results using the BPA concentrations continuously. The associations were also linear for all other continuous covariates (P-gain >0.10), except for maternal prepregnancy BMI (P-gain = 0.05); log transformation gave a linear association, and therefore, we adjusted multivariable models for the log-transformed maternal prepregnancy BMI variable.
For every outcome variable, we first studied the association of interest in the crude model. We then constructed a basic model adjusted for child sex, exact age at which the outcome variable was measured, and exact time of day the urine samples were collected. We evaluated potential confounding by other covariates using a forward-selection approach, adding each covariate one at the time to the basic model. We retained in the final multivariable-adjusted models only those covariates that modified the coefficient of the BPA exposure covariate in the basic model by >5%. Because little is known about the factors that are associated with BPA exposure levels in the population, we evaluated a large number of covariates as potential confounders. First, we selected candidate confounders based on our analyses of determinants of BPA exposure26: maternal age at delivery, maternal education (primary, secondary, university), parity (nulliparous, multiparous), maternal smoking during pregnancy (no/yes), and 1st or 3rd trimester dietary intakes of canned fish (<0.9, 0.9–1.9, ≥2 servings per week). Furthermore, we selected other potential confounders based on the literature: child’s birth length and height at age 14 months and 4 years, gestational age, exclusive breastfeeding duration (<16, ≥16 weeks), maternal country of origin (Spain, other), maternal prepregnancy BMI status (log transformed), maternal weight gain during pregnancy,28 maternal social class (I and II: professionals and managers, III: other nonmanual, IV and V: skilled, semiskilled, and unskilled manual), maternal DDE serum concentrations in pregnancy (because DDE has been related to postnatal growth in this cohort),29 and paternal education, social class, and BMI. In the models assessing the associations at 4 years of age, we additionally evaluated as confounders the child’s consumption of fast food and sugar-sweetened beverages and time spent watching TV or playing videogames at that age. In the analysis population, no mothers had reported gestational diabetes or any other metabolic diseases that could have altered BPA secretion in urine and may be related to child postnatal growth.
Following these methods, all final models were adjusted for child’s sex, exact age at the time the outcome was measured, exact time of day of urine sample collection, and the following maternal characteristics: country of origin, age at delivery, education, parity, prepregnancy BMI, and smoking during pregnancy. Models for waist circumference Z score were additionally adjusted for child’s height. To evaluate the robustness of results, we stratified analyses by the main covariates (child’s sex, exclusive breastfeeding duration, and maternal characteristics [age at delivery, education, and smoking during pregnancy]), and we performed likelihood ratio tests of their interaction terms.
All analyses for the outcomes assessed in the first 6 months of life and at age 14 months were conducted in the sample with complete data on average BPA concentrations, outcome, and other covariates included in the models (n = 402; data on other covariates were missing for less than 1% [n = 2] of the mother–child pairs with complete BPA and outcome data). The models of BMI Z score and overweight at age 4 years included fewer observations (n = 348) because of missing values in the outcome variable (n = 54). There were four more infants with missing values for waist circumference at the same age.
As a sensitivity analysis we repeated all analyses using (1) the 1st and 3rd trimester average BPA concentration without creatinine adjustment (in μg/L) and (2) the 1st and 3rd trimester average BPA concentration (in μg/L), further adding 1st and 3rd trimester creatinine concentrations in the model as separate covariates. Results did not change meaningfully, and thus, we present here only the results using creatinine-adjusted concentrations. Furthermore, we carried out sensitivity analyses of the associations between the two single-spot BPA measurements separately and the obesity-related outcomes. We repeated all analyses excluding children of mothers with urinary creatinine values <0.2 g/L at either 1st (n = 8) or 3rd (n = 5) trimesters of pregnancy, to reduce the influence of very dilute specimens.30 We also restricted analysis to the 344 children with complete BPA and growth data up to 4 years of age. Finally, we repeated analyses including preterm births. All statistical analyses were performed using the statistical package STATA 10.1 (Stata Corporation, College Station, TX).
Twenty-six percent of the children included in the analysis were rapid growers in the first 6 months of life (Table 1). The prevalence of overweight was 25% at age 14 months and 21% at age 4 years. Waist circumference and BMI Z score measurements were correlated at both age 14 months and 4 years (Pearson r = 0.77 and 0.79, respectively). Mothers of children included in the analysis were of higher education and were less likely to have smoked during pregnancy or to report gestational diabetes compared with mothers of excluded children (eAppendix, eTable 1, http://links.lww.com/EDE/A706).
In the analysis population, the population geometric means of the creatinine-adjusted BPA concentrations were 2.6 μg/g in the 1st trimester and 2.0 μg/g in the 3rd trimester of pregnancy, and the average was 2.6 μg/g (Table 2). Creatinine concentrations were somewhat higher in the 3rd compared with the 1st trimester (geometric mean = 0.9 and 0.8 g/L, respectively). Log10-transformed creatinine-adjusted BPA concentrations were weakly correlated between the 1st and 3rd trimesters (Pearson correlation coefficient r = 0.13). BPA average concentrations were higher in women of younger age of lower education and in those who were nulliparous or smoked during pregnancy (Table 3). BPA average concentrations did not vary according to other characteristics.
The average of the 1st and 3rd trimester creatinine-adjusted BPA concentrations was not associated with weight gain Z score or rapid growth in the first 6 months of life (Table 4). At age 14 months, no association was found between the average BPA concentration and waist circumference Z score, BMI Z score, or overweight risk. Effect estimates of the crude models for the associations between BPA and all the obesity-related outcomes under study did not change substantially in the final models adjusted for child and maternal characteristics. BPA associations with the outcomes assessed in the first 6 months of life and at the age of 14 months were homogeneous in the subgroups defined by child sex, exclusive breastfeeding duration, or maternal characteristics (age at delivery, education, and smoking in pregnancy) (not shown).
At 4 years of age, the average BPA concentration was weakly associated with an increase in the waist circumference Z score (adjusted β per log10 μg BPA/g creatinine = 0.28 [95% confidence interval (CI) = 0.01 to 0.57]) (Table 4 and Figure A). Associations were stronger among children of mothers who had smoked during pregnancy compared with children of mothers who had not (adjusted β = 0.66 [95% CI = 0.08 to 1.30] and 0.08 [−0.26 to 0.43] respectively; test for interaction, P = 0.08). Associations did not vary by other covariates (not shown). BPA was also weakly associated with the BMI Z score at 4 years of age (adjusted β = 0.28 [95% CI = −0.06 to 0.63]) (Table 4 and Figure B). The association was somewhat stronger among children of mothers who had smoked during pregnancy compared with children of mothers who had not (test for interaction, P = 0.26). There was weak evidence for BPA exposure to be associated with an increased risk of overweight at 4 years (adjusted RR = 1.38 [95% CI = 0.72 to 2.67]) (Table 4). Excluding outliers (two children with the highest and one with the lowest average BPA concentrations) resulted in increased effect estimates for both waist circumference and BMI Z scores (adjusted β = 0.35 [95% CI = 0.04 to 0.66] and 0.41 [0.03 to 0.79], respectively) and the risk of overweight (adjusted RR = 1.70 [95% CI = 0.90 to 3.45]).
In sensitivity analyses separately using the single-spot BPA concentrations measured in the 1st and 3rd trimester, there was no indication that 1st- or 3rd- trimester BPA concentrations were more strongly associated with the obesity-related outcomes (eTables 2 and 3, http://links.lww.com/EDE/A706). The β coefficients for the associations between BPA and waist circumference and BMI Z scores at 4 years of age were in the same direction for the single-spot BPA concentrations compared with the average. Excluding children of mothers with urinary creatinine values <0.2 g/L at either 1st or 3rd trimesters of pregnancy did not influence results (not shown). After restricting the analysis to children with complete follow-up data through the age of 4 years (n = 344), results related to the outcomes measured at previous ages did not change meaningfully (not shown). We obtained similar effect estimates when including preterm births and when using weight-for-length age- and sex-specific Z scores and waist-to-height ratio, instead of, respectively, BMI Z scores and waist circumference Z scores (not shown).
Prenatal BPA exposure was weakly associated with increased waist circumference and BMI in children at the age of 4 years, but not with weight gain and rapid growth in the first 6 months of life, or with waist circumference or BMI at 14 months of age. Associations were somewhat stronger in children of mothers who smoked during pregnancy.
The urinary mean BPA concentrations measured in this study are similar to those reported previously in pregnant women of other European7,31 or American6,32 populations, but higher than those reported recently in a Latina population by Harley and colleagues.13 The recent cross-sectional study using data from the 2003 to 2008 National Health and Nutrition Examination Survey (NHANES) suggested a positive association between urinary BPA concentrations and elevated BMI in white children and adolescents aged 6 to 19 years, but not in blacks or Hispanics.12 Harley et al found in cross-sectional analyses a positive association of child’s urine BPA concentrations with BMI, waist circumference, and fat mass at the age of 9 years in a US Latina population. However, in that same population, maternal urine BPA concentrations during pregnancy (also using the average of two spot-urine measurements) were negatively associated among girls with BMI Z scores from 2 to 9 years of age and with fat mass at the age of 9 years, whereas no effect was observed in boys.13 Continued follow-up will be needed to evaluate our observed effects at prepubertal and adolescent ages.
Our study suggests that prenatal BPA exposure at levels detected in the general population may have a weak obesogenic effect that cannot be detected in anthropometric measurements in early infancy, but that may become apparent after 4 years of age. Nevertheless, methodological aspects may also explain the findings of this study. Urinary BPA concentrations between the 1st and 3rd trimesters of pregnancy were only weakly correlated. BPA has a very short-half life in human biological tissues,33 and urinary BPA concentrations have been shown to present an important degree of within-subject variability in populations5 similar to ours. In our study, the within-individual variability (0.36) was far greater than between-individual variability (SD = 0.08; intraclass correlation coefficient = 0.05 [95% CI = 0.00 to 0.17]). Thus, even though we have used the average of two spot-urine concentrations as our best proxy of BPA exposure throughout pregnancy (a great advantage over using just one spot-urine test), it is likely that exposure misclassification has biased our results toward the null. Nevertheless, when we classified subjects into categories of “consistently low” (both spot measurements in the lowest exposure tertile or one in the lowest and one in the middle), “consistently high” (both spot measurements in the highest tertile or one in the highest and one in the middle, and “medium” (all others), we found good agreement with a classification according to tertiles of the average BPA concentration (Cohen’s weighted κ = 0.70); findings for the obesity-related outcomes were similar, with indications for associations only at age 4 (not shown).
Obesity in infancy has been shown to track into later childhood (especially after the age of 2 years and into adulthood).34 Although prenatal BPA exposure was not clearly related to overweight at the age of 4 years (potentially due to sample size restrictions), BPA was associated with both BMI and waist circumference at this age. The fact that results using BMI and using waist circumference as indicators of childhood obesity were consistent strengthens confidence in our findings. In a recent study, overweight/obese school-age children and adolescents followed an annual hospital or public healthcare treatment with the aim of reducing their BMIs. This study suggested that even small reductions in BMI Z scores (<0.1 SDs) are associated with an improvement in cardiovascular risk factors including a reduction in insulin and low-density lipoprotein.35 Thus, the BMI increases observed in our study (ie, mean BMI Z score increase equivalent to 0.22 SDs [95% CI = −0.05 to 0.50] in children in the highest compared with the lowest tertile of BPA exposure) could be an important indicator of physiologic changes related to later disease risk. A potential limitation of this study is the fact that we have information only for indirect indicators of adiposity (ie, weight gain, BMI, and waist circumference). The use of more direct measurements of body composition that permit distinguishing fat mass from lean mass (eg, bioelectrical impedance and dual-energy X-ray absorptiometry), or the measurement of lipid and hormone biomarkers related to metabolic syndrome, would be important in future studies.
The “environmental obesogen hypothesis” postulates that chemicals with endocrine-disrupting properties may perturb nuclear-hormone-receptor signaling in preadipocytes and mature adipocytes and may alter adipogenic gene expression, promoting adipocyte differentiation and fatty acid storage.36 In vitro and in vivo studies have shown that BPA may act as an obesogen.37 BPA exposure has been shown to induce adipogenesis and triglyceride accumulation in mouse 3T3-L1 preadipocytes38 and to inhibit the release of adiponectin from mature human adipocytes.39Adiponectin is a hormone highly expressed in adipose tissue and thought to protect against insulin resistance and metabolic syndrome.40 The obesogenic effects of perinatal BPA exposure at low doses have been further suggested to be enhanced and more persistent in female rats.8,10 In this study, we did not find evidence that child’s sex modifies the association between prenatal BPA exposure and the obesity-related outcomes under study. Effect heterogeneity by child sex should be further explored in larger populations.
A main strength of our study is that its prospective study design rules out reverse causality, a main weakness in the existing literature. The prospective design has enabled us to obtain repeated anthropometric measurements and to evaluate the association of prenatal BPA exposure with postnatal growth predictors at various ages. Our effect estimates did not change substantially after inclusion of a wide range of child and maternal characteristics as potential confounders, and effect estimates were relatively robust in population subgroups defined by main covariates with the exception of smoking. We found evidence that some risk factors (especially social class and maternal prepregnancy BMI) influenced the outcomes at age 4 years more than the outcomes at younger ages (not shown), indicating that the influence of unmeasured confounding may also be stronger at later ages. On the other hand, few of the measured confounders (including social class, dietary and sedentary activity factors) influenced the effect estimates at 4 years, and it is possible that a true effect of BPA would not become apparent until later age (eg, through epigenetic modifications that affect the phenotype later in childhood). Maternal tobacco smoke is a relatively well-established obesogen,2 and the possibility of an interaction effect between smoking and BPA warrants further investigation. In a subgroup of 113 children from this cohort study, mean urinary BPA concentrations measured at age 4 years were not correlated with the maternal BPA concentrations measured in pregnancy (P > 0.10), and thus postnatal BPA exposure is unlikely to have confounded the associations shown here. Finally, selection bias is a potential limitation of our study. Compared with children excluded from analysis, those included were more likely to have mothers with higher education and less likely to have mothers who smoked during pregnancy or who reported gestational diabetes. Lower maternal education and smoking during pregnancy are associated with increased maternal BPA concentrations in this cohort and overweight in infancy in this population. The potential exclusion from the analysis of the highest exposed children, who might be more susceptible to the effects under study, may have led to an underestimation of risk estimates in our study.
In conclusion, this study provides some evidence for an association between BPA exposure and obesity-related outcomes at the age of 4 years, although not in infancy. The short half-life and large within-person variability of BPA exposure create large uncertainties in exposure assessment. Repeated measurements of BPA concentrations at many time points, prenatally and also postnatally, would improve population exposure classification and further could permit the evaluation of the most important windows of susceptibility for the BPA effects (if any) on childhood obesity.
We are grateful to Xavier Basagaña for providing statistical advice and to all the cohort participants for their generous collaboration. A full roster of the INMA Project Investigators can be found at: http://www.proyectoinma.org.
1. Barouki R, Gluckman PD, Grandjean P, Hanson M, Heindel JJ. Developmental origins of non-communicable disease: implications for research and public health. Environ Health. 2012;11:42
2. Oken E, Levitan EB, Gillman MW. Maternal smoking during pregnancy and child overweight: systematic review and meta-analysis. Int J Obes (Lond). 2008;32:201–210
3. Thayer KA, Heindel JJ, Bucher JR, Gallo MA. Role of environmental chemicals in diabetes and obesity: a National Toxicology Program workshop review. Environ Health Perspect. 2012;120:779–789
4. Vandenberg LN, Hauser R, Marcus M, Olea N, Welshons WV. Human exposure to bisphenol A (BPA). Reprod Toxicol. 2007;24:139–177
5. Braun JM, Kalkbrenner AE, Calafat AM, et al. Variability and predictors of urinary bisphenol A concentrations during pregnancy. Environ Health Perspect. 2011;119:131–137
6. Calafat AM, Ye X, Wong LY, Reidy JA, Needham LL. Exposure of the U.S. population to bisphenol A and 4-tertiary-octylphenol: 2003-2004. Environ Health Perspect. 2008;116:39–44
7. Ye X, Pierik FH, Hauser R, et al. Urinary metabolite concentrations of organophosphorous pesticides, bisphenol A, and phthalates among pregnant women in Rotterdam, the Netherlands: the Generation R study. Environ Res. 2008;108:260–267
8. Rubin BS, Murray MK, Damassa DA, King JC, Soto AM. Perinatal exposure to low doses of bisphenol A affects body weight, patterns of estrous cyclicity, and plasma LH levels. Environ Health Perspect. 2001;109:675–680
9. Rubin BS, Soto AM. Bisphenol A: Perinatal exposure and body weight. Mol Cell Endocrinol. 2009;304:55–62
10. Somm E, Schwitzgebel VM, Toulotte A, et al. Perinatal exposure to bisphenol a alters early adipogenesis in the rat. Environ Health Perspect. 2009;117:1549–1555
11. Vandenberg LN, Colborn T, Hayes TB, et al. Hormones and endocrine-disrupting chemicals: low-dose effects and nonmonotonic dose responses. Endocr Rev. 2012;33:378–455
12. Trasande L, Attina TM, Blustein J. Association between urinary bisphenol A concentration and obesity prevalence in children and adolescents. JAMA. 2012;308:1113–1121
13. Harley KG, Aguilar Schall R, Chevrier J, et al. Prenatal and postnatal bisphenol A exposure and body mass index in childhood in the CHAMACOS cohort. Environ Health Perspect. 2013;121:514–520, 520e1
14. Druet C, Stettler N, Sharp S, et al. Prediction of childhood obesity by infancy weight gain: an individual-level meta-analysis. Paediatr Perinat Epidemiol. 2012;26:19–26
15. Stettler N. Nature and strength of epidemiological evidence for origins of childhood and adulthood obesity in the first year of life. Int J Obes (Lond). 2007;31:1035–1043
16. Monteiro PO, Victora CG. Rapid growth in infancy and childhood and obesity in later life–a systematic review. Obes Rev. 2005;6:143–154
17. Guxens M, Ballester F, Espada M, et al.INMA Project. Cohort Profile: the INMA–INfancia y Medio Ambiente–(Environment and Childhood) Project. Int J Epidemiol. 2012;41:930–940
18. Johnson MJ, Wootton SA, Leaf AA, Jackson AA. Preterm birth and body composition at term equivalent age: a systematic review and meta-analysis. Pediatrics. 2012;130:e640–e649
20. Willett W, Stampfer MJ. Total energy intake: implications for epidemiologic analyses. Am J Epidemiol. 1986;124:17–27
21. Guxens M, Mendez MA, Moltó-Puigmartí C, et al. Breastfeeding, long-chain polyunsaturated fatty acids in colostrum, and infant mental development. Pediatrics. 2011;128:e880–e889
23. Simondon KB, Simondon F, Delpeuch F, Cornu A. Comparative study of five growth models applied to weight data from congolese infants between birth and 13 months of age. Am J Hum Biol. 1992;4:327–335
24. de Onis M, Garza C, Onyango AW, Rolland-Cachera MF. WHO growth standards for infants and young children [in French]. Arch Pediatr. 2009;16:47–53
25. McCarthy HD, Ashwell M. A study of central fatness using waist-to-height ratios in UK children and adolescents over two decades supports the simple message–‘keep your waist circumference to less than half your height’. Int J Obes (Lond). 2006;30:988–992
26. Casas M, Valvi D, Luque N, et al. Dietary and sociodemographic determinants of bisphenol A urine concentrations in pregnant women and children. Environ Int. 2013;56:10–18
27. Mahalingaiah S, Meeker JD, Pearson KR, et al. Temporal variability and predictors of urinary bisphenol A concentrations in men and women. Environ Health Perspect. 2008;116:173–178
28. Casas M, Chatzi L, Carsin AE, et al. Maternalpre-pregnancy overweight and obesity, and child neuropsychological development: two Southern European birth cohort studies. Int J Epidemiol. 2013;42:506–517
29. Mendez MA, Garcia-Esteban R, Guxens M, et al. Prenatal organochlorine compound exposure, rapid weight gain, and overweight in infancy. Environ Health Perspect. 2011;119:272–278
30. Wolff MS, Engel SM, Berkowitz GS, et al. Prenatal phenol and phthalate exposures and birth outcomes. Environ Health Perspect. 2008;116:1092–1097
31. Philippat C, Mortamais M, Chevrier C, et al. Exposure to phthalates and phenols during pregnancy and offspring size at birth. Environ Health Perspect. 2012;120:464–470
32. Braun JM, Yolton K, Dietrich KN, et al. Prenatal bisphenol A exposure and early childhood behavior. Environ Health Perspect. 2009;117:1945–1952
33. Hengstler JG, Foth H, Gebel T, et al. Critical evaluation of key evidence on the human health hazards of exposure to bisphenol A. Crit Rev Toxicol. 2011;41:263–291
34. Brisbois TD, Farmer AP, McCargar LJ. Early markers of adult obesity: a review. Obes Rev. 2012;13:347–367
35. Kolsgaard ML, Joner G, Brunborg C, Anderssen SA, Tonstad S, Andersen LF. Correction: “Reduction in BMI z-score and improvement in cardiometabolic risk factors in obese children and adolescents. The Oslo adiposity intervention study - a hospital/public health nurse combined treatment.”. BMC Pediatr. 2012;12:77
36. Janesick A, Blumberg B. Obesogens, stem cells and the developmental programming of obesity. Int J Androl. 2012;35:437–448
37. Vom Saal FS, Nagel SC, Coe BL, Angle BM, Taylor JA. The estrogenic endocrine disrupting chemical bisphenol A (BPA) and obesity. Mol Cell Endocrinol. 2012;354:74–84
38. Masuno H, Iwanami J, Kidani T, Sakayama K, Honda K. Bisphenol a accelerates terminal differentiation of 3T3-L1 cells into adipocytes through the phosphatidylinositol 3-kinase pathway. Toxicol Sci. 2005;84:319–327
39. Hugo ER, Brandebourg TD, Woo JG, Loftus J, Alexander JW, Ben-Jonathan N. Bisphenol A at environmentally relevant doses inhibits adiponectin release from human adipose tissue explants and adipocytes. Environ Health Perspect. 2008;116:1642–1647
40. Whitehead JP, Richards AA, Hickman IJ, Macdonald GA, Prins JB. Adiponectin–a key adipokine in the metabolic syndrome. Diabetes Obes Metab. 2006;8:264–280
Supplemental Digital Content
Copyright © 2013 Wolters Kluwer Health, Inc. All rights reserved.