Abstract
Background
Bisphenol A (BPA) is an environmental estrogen used in the manufacture of polycarbonate plastics and epoxy resins used to make food and beverage packaging. Increasing evidence suggests that BPA mimics estrogens in the body and may be associated with putative markers of breast cancer risk.
Objectives
We analyzed the National Health and Nutrition Examination Survey (NHANES) 2003–2010 data to investigate the association of BPA with age at menarche in adolescent girls. We hypothesized that urinary BPA, as a surrogate biomarker for BPA exposure, is associated with earlier age at menarche, and that body mass index (BMI) may modulate this association.
Methods
We conducted cross-sectional analyses of urinary BPA, BMI and age of menarche in a subsample of 987 adolescent girls aged 12–19, using pooled data from the 2003–2010 NHANES. Unconditional logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CI) for the association between urinary BPA and early onset of menarche, with adjustment for sampling design. We additionally assessed interaction of BPA with BMI.
Results
Adolescent girls with moderate BPA levels appeared to be less likely to have early onset of menarche than those with the lowest levels (OR=0.57; 95% CI=0.30, 1.08) after adjusting for age, race/ethnicity, parental education, country of birth, NHANES cycle, BMI and creatinine. BMI appeared to modify the BPA-menarche association.
Conclusions
Although a non-significant trend suggests increasing urinary BPA may be associated with delayed menarche in adolescent girls, these results are based on cross-sectional data. Results should be clarified in carefully designed longitudinal cohort studies.
Keywords: puberty, Bisphenol A, endocrine disruptors, breast cancer, menarche
1. Introduction
Bisphenol A (BPA) is an estrogen-like endocrine disrupting chemical (EDC) of phenolic nature that has been used for over 50 years in the manufacture of polycarbonate plastics and epoxy resins (Centers for Disease Control and Prevention, 2012). Polycarbonate plastics are used in food and drink packaging, while epoxy resins are used to coat the inside of metal cans. BPA can leach into food sources, particularly at high temperatures, thus dietary intake is suggested to be the main source of BPA exposure for the majority of the population (von Goetz et al., 2010). The Environmental Protection Agency (EPA) considers BPA a high production chemical, with the United States (US) volume of BPA totaling 2.4 billion pounds in 2007 (U.S. Environmental Protection Agency, 2010). Thus, due to its high volume production and ubiquitous application, human exposure to BPA is widespread. Data from the 2003–2004 National Health and Nutrition Examination Survey (NHANES) revealed detectable levels of BPA in 92.6% of individuals 6 years of age and older (Calafat et al., 2008).
There is little conclusive evidence of the health effects from BPA exposure in humans at low environmental doses. In animal studies, high dose BPA exposure has been shown to disrupt reproductive development including preputial separation (Ema et al., 2001; Tyl et al., 2002), delayed vaginal opening (Ashby and Tinwell, 1998) and delayed testicular descent (Nagao et al., 1999). Additionally, BPA exposure has demonstrated neurological effects including stimulation of neuronal differentiation and possible disruption of neonatal brain development (Kim et al., 2009). Animal studies conducted in rats suggest possible effects such as disruption to the hypothalamic-pituitary-gonadal axis (Rasier et al., 2006), changes to fetal mammary gland morphology (Moral et al., 2008), and carcinogenesis (Betancourt et al., 2010; Jenkins et al., 2009).
The adverse effects of EDCs such as BPA are especially prominent during puberty, with evidence from both animal and human studies (Parent et al., 2005). It has been suggested that at the onset of puberty, the neuroendocrine processes become highly vulnerable to environmental factors that can permanently affect the development and functionality of reproductive organs, growth spurt, and maturation of the brain (Roy et al., 2009). Epidemiological studies have shown associations with EDCs such as phthalates, dichlorodiphenyldichloroethylene (DDE), polybrominated biphenyls (PBB) and early onset of pubertal development in girls (Ozen and Darcan, 2011). BPA specifically has been shown to act as a hormonal agonist in animal models of reproductive development, thus accelerating pubertal development (Honma et al., 2002).
The average age of pubertal onset in girls has fallen over the years (Biro et al., 2010). Studies also suggest a decline in the age of onset of menarche in girls (Herman-Giddens, 2006; Kaplowitz, 2008). Early age of menarche in girls is associated with many adverse psychosocial and health outcomes later in life including obesity, depression, and breast cancer (Black and Klein, 2012). Specifically, research supports about a 30% increase in the risk of breast cancer for early age of menarche (Biro and Wolff, 2011). Of increasing concern is the association between environmental exposures and early menarche, particularly for EDCs such as BPA (Biro et al., 2009).
There is very limited epidemiologic data on BPA exposure in children as it relates to pubertal development. The current study analyzed data from the 2003–2010 NHANES to investigate the association of urinary BPA with age at menarche in adolescent girls. We hypothesized that urinary BPA concentration is associated with earlier age at menarche. Additionally, obesity is a strong endogenous hormonal risk factor for pubertal development (Kaplowitz, 2008). We hypothesized that obesity, as measured by body mass index (BMI), may modify the association between urinary BPA concentrations and age at menarche.
2. Methods
2.1. Study population
Pooled data from the 2003–2004, 2005–2006, 2007–2008, and 2009–2010 NHANES was used to investigate the association between urinary BPA concentrations and age at menarche in adolescent girls. NHANES utilizes a complex, multistage, probability cluster design of the civilian, non-institutionalized US population. Four sets of NHANES data were selected (2003–2010) and merged in order to create more precise survey estimates. The sample was restricted to female adolescents (aged 12–19 years) in order to assess BPA exposure near the age of menarche (n=3,493). Female adolescents were included if they had completed the reproductive health questionnaire and had complete data on age at menarche (n=3,114). Urinary BPA concentrations were measured in a random, one-third subsample of all participants aged 6 years and up. The study sample was further restricted to those that had measurements of BPA. The final sample size consisted of 987 adolescent girls.
2.2. Urinary BPA concentrations
One spot urine samples were collected from participants in the 2003–2010 continuous NHANES. Urinary BPA concentrations were measured using online solid phase extraction (SPE) coupled with high performance liquid chromatography (HPLC) isotope dilution-tandem mass spectrometry (MS/MS). Analysis was conducted at the Division of Environmental Health Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention (CDC) (National Center for Health Statistics, 2010). Coefficients of variation for low and high-concentrations were 18.6% and 12.1% in 2003–2004, 12.6% and 11.2% in 2005–2006, 8.1% and 5.7% in 2007–2008, and 8.1% and 5.7% in 2009–2010, respectively. The lower limit of detection (LLOD) was uniform across all years of data collection: 0.36 in 2003–2004, 0.4 in 2005–2006, 0.4 in 2007–2008, and 0.4 in 2009–2010. We categorized urinary BPA concentrations into tertiles based on the distribution of BPA in girls who achieved menarche at or above age 12.
2.3. Age at menarche
Age at menarche was assessed using the reproductive health questionnaire by asking girls aged ≥16 years “How old were you when you had your first menstrual period?” For girls aged 12–15, their parent or guardian was used as a proxy and completed the questionnaire for them.
Of the 987 adolescent girls, 59 had not reached menarche prior to urinary analysis. Of the 59 girls who had not reached menarche, 58 (98%) of them were aged 12–15. Age of menarche was categorized as <12 and ≥12 years of age based on the mean age among girls who attained menarche. Menarche was considered “early” if it was achieved before 12 years of age.
2.4. Potential confounders
Potential confounders were identified from the literature and include: age (continuous in years), race/ethnicity (non-Hispanic white, non-Hispanic black, Mexican American, other/multi-racial), poverty income ratio (PIR: ratio of income to the family’s appropriate poverty threshold, categorized as <1, 1–2, 2–3, and ≥3), country of birth (assessed by asking “In what country were you born”, categorized as U.S. born or not), level of guardian education (<high school, high school or equivalent, and college graduate), annual household income (<$20,000, $20,000 to $35,000, $35,000 to $65,000, and ≥$65,000), and body mass index (BMI). Centers for Disease Control and Prevention (CDC) 2000 growth charts were used to convert NHANES BMI values into age- and gender-specific BMI percentile levels (<50th, 50–85th, and ≥85th percentile).
2.5. Statistical analysis
In order to account for the complex sampling design used in NHANES, all analyses were adjusted for sampling strata and clusters. Corresponding weights for each round of the sample were used. Weighted estimates were computed in accordance with the NHANES Analytic and Reporting Guidelines, and the smallest subsample weights were used (urinary BPA subsamples B and C) (National Center for Health Statistics, 2006). Taylor series linearization was used to take into account clustering using the masked variance pseudo-psu and pseudo-stratum variables.
Logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for the association between urinary BPA and menarche before 12 years of age. We further examined if this association was modified by BMI by including an interaction term between urinary BPA and BMI and by using the Wald test. BMI was categorized as overweight (≥85th percentile) and not overweight (<85th percentile) for the interaction analyses. Chi-square tests were used to assess associations between urinary BPA tertiles and potential cofounders. If a covariate from the list of potential confounders changed the OR between tertiles of urinary BPA and age of menarche by 10 percent or more in a logistic regression model, then it was included in the final model as a confounder. Urinary BPA concentrations were positively skewed, thus, log-transformation was applied to the measurements to achieve a normal distribution. Linear trend of BPA exposure and age at menarche was assessed by using the log-transformed values of urinary BPA. The values presented in the tables were reverse-log-transformed. Statistical significance was defined by p-values <0.05. All statistical analyses were performed using Stata 11.0 (Stata Corp, LP, College Station, TX). Finally, all models were adjusted for urinary creatinine to correct for urinary dilution (Barr et al., 2005).
3. Results
The study sample comprised of 987 adolescent girls aged 12–19 years in the 2003–2010 NHANES. The weighted study sample was primarily non-Hispanic white (62.5%), from a household that had an annual income greater than $65,000 (40.3%), and had parents with more than a high school degree (54.1%). There was a higher percentage of girls in the 50–85th (37.3%) and ≥85th (35.3%) BMI age- and sex-specific percentiles (Table 1). Higher urinary BPA concentrations were observed for non-Hispanic Whites, African Americans, and those in the ≥85th BMI percentile. Girls in the 12–15 year old age category had similar BPA concentrations compared to those in the 16–19 age category. PIR, annual household income, and level of guardian education were not associated with urinary BPA concentrations (Table 1).
Table 1.
Sample characteristics across urinary BPA tertiles in 2003–2010 NHANES
| Total N (%)a |
Urinary BPA concentrations (ng/mL)
|
p-valueb | |||
|---|---|---|---|---|---|
| Low (<1.9) (%) | Moderate (1.9–4.1) (%) | High (>4.1) (%) | |||
| Total | 987 | 336 (34.0) | 324 (32.8) | 327 (33.1) | |
| Age (years) | |||||
| 12–15 | 482 (49.0) | 164 (33.8) | 169 (35.7) | 149 (30.5) | 0.77 |
| 16–19 | 505 (51.0) | 172 (35.4) | 155 (32.9) | 178 (31.7) | |
| Race/ethnicity | |||||
| Non-Hispanic Whites | 293 (62.5) | 95 (33.8) | 96 (35.0) | 102 (31.2) | 0.03 |
| African Americans | 296 (14.7) | 69 (25.6) | 95 (34.3) | 132 (40.1) | |
| Mexican Americans | 278 (11.2) | 125 (43.4) | 94 (36.6) | 59 (20.1) | |
| Other | 120 (11.6) | 47 (41.9) | 39 (27.7) | 34 (30.4) | |
| Annual household income | |||||
| <$20,000 | 229 (15.6) | 77 (33.1) | 65 (28.7) | 87 (38.3) | 0.74 |
| $20,000–$35,000 | 225 (18.2) | 76 (34.2) | 74 (35.2) | 75 (30.6) | |
| $35,000–$65,000 | 234 (21.3) | 77 (33.1) | 86 (36.4) | 71 (30.5) | |
| ≥ $65,000 | 259 (40.3) | 88 (35.1) | 88 (36.0) | 83 (29.0) | |
| Missing | 40 (4.6) | 18 (44.5) | 11 (24.5) | 11 (31.0) | |
| Poverty income ratio | |||||
| <1.0 | 299 (21.8) | 96 (33.0) | 99 (35.3) | 104 (31.8) | 0.24 |
| 1.0–1.9 | 258 (20.0) | 92 (34.8) | 80 (34.8) | 86 (30.5) | |
| 2.0–2.9 | 134 (15.1) | 36 (23.9) | 52 (37.1) | 46 (39.1) | |
| ≥3 | 246 (37.7) | 93 (39.2) | 78 (33.9) | 75 (26.8) | |
| Missing | 50 (5.5) | 19 (38.2) | 15 (23.0) | 16 (38.8) | |
| U.S. born | |||||
| Yes | 862 (91.4) | 277 (33.1) | 293 (35.3) | 292 (31.6) | 0.01 |
| No | 125 (8.6) | 59 (50.5) | 31 (23,2) | 35 (26.3) | |
| Level of guardian education | |||||
| <High school | 311 (20.2) | 101 (35.1) | 102 (28.3) | 108 (36.6) | 0.39 |
| High school or equivalent | 215 (20.6) | 76 (33.2) | 64 (32.0) | 75 (34.8) | |
| >High school | 415 (54.1) | 139 (33.6) | 142 (37.2) | 134 (29.3) | |
| Missing | 46 (5.1) | 20 (49.4) | 16 (35.9) | 10 (14.8) | |
| BMI age- and sex-specific percentile | |||||
| <50th | 238 (27.4) | 91 (35.8) | 73 (35.5) | 74 (28.7) | 0.78 |
| 50–85th | 349 (37.3) | 132 (36.6) | 107 (32.7) | 110 (30.7) | |
| ≥85th | 400 (35.3) | 113 (31.6) | 144 (34.9) | 143 (33.5) | |
Abbreviations: BPA, bisphenol A; BMI, body mass index (kg/m2).
Percentages are weighted using population weights for the BPA subsample.
Calculated using a Chi-square test.
The overall geometric mean of urinary BPA concentration was 2.64 ng/mL (95% CI 2.41, 2.89). The geometric means of urinary BPA were 3.49 ng/mL (95% CI 2.86, 4.26) in the 2003–2004 NHANES cycle, 2.39 ng/mL (95% CI 2.15, 2.66) in the 2005–2006 cycle, 2.78 ng/mL (95% CI 2.19, 3.52) in the 2007–2008 cycle, and 2.08 ng/mL (95% CI 1.75, 2.47) in the 2009–2010 cycle (data not shown).
Table 2 shows the mean age at menarche and percentage of menarche before 12 years of age. The mean age of menarche was 12.14 years and ranged from 8 to 17 years among girls who had already attained menarche. Overall, 27.8% of girls had obtained menarche before 12 years of age. Among girls who experienced menarche, mean age of menarche differed by age group at time of survey, race/ethnicity, and BMI. Considering all girls in the sample, the percentage of girls who experienced menarche before age 12 also differed by age at time of survey, race/ethnicity and BMI. No other differences were observed between potential confounders and age of menarche.
Table 2.
Mean age at menarche and percentage of menarche <12 years of age, NHANES, 2003–2010
| N | Mean menarche age (95% CI)a | N | Menarche <12 years (%)b | |
|---|---|---|---|---|
| Total | 928 | 12.14 (12.04, 12.24) | 987 | 301 (27.8) |
| Urinary BPA concentrations (ng/mL) | ||||
| Low (<1.9) | 319 | 12.06 (11.85, 12.26) | 336 | 111 (32.5) |
| Moderate (1.9–4.1) | 316 | 12.34 (12.20, 12.47) | 324 | 97 (23.1) |
| High (>4.1) | 293 | 12.02 (11.82, 12.22) | 327 | 93 (27.8) |
| Age (years) # | ||||
| 12–15 | 424 | 11.75 (11.63, 11.87) | 482 | 166 (32.1) |
| 16–19 | 504 | 12.47 (12.33, 12.61) | 505 | 135 (23.7) |
| Race/ethnicity# | ||||
| Non-Hispanic Whites | 272 | 12.28 (12.12, 12.44) | 293 | 68 (23.3) |
| African Americans | 280 | 11.95 (11.75, 12.16) | 296 | 110 (38.6) |
| Mexican Americans | 262 | 11.91 (11.75, 12.07) | 278 | 85 (34.6) |
| Other | 114 | 11.88 (11.60, 12.17) | 120 | 38 (32.2) |
| Annual household income | ||||
| <$20,000 | 223 | 12.24 (11.99, 12.49) | 229 | 64 (26.4) |
| $20,000–$35,000 | 212 | 12.12 (11.91, 12.33) | 225 | 73 (30.7) |
| $35,000–$65,000 | 214 | 12.03 (11.84, 12.21) | 234 | 67 (24.6) |
| ≥ $65,000 | 240 | 12.21 (12.04, 12.38) | 259 | 83 (27.7) |
| Missing | 39 | 40 | ||
| Poverty income ratio | ||||
| <1 | 289 | 12.13 (11.90, 12.35) | 299 | 90 (29.5) |
| 1- | 237 | 12.20 (11.99, 12.41) | 258 | 71 (24.1) |
| 2- | 123 | 11.89 (11.59, 12.19) | 134 | 42 (32.0) |
| ≥3 | 230 | 12.27 (12.09, 12.45) | 246 | 79 (25.5) |
| Missing | 49 | 50 | ||
| U.S. born | ||||
| Yes | 807 | 12.14 (12.04, 12.25) | 862 | 269 (27.5) |
| No | 121 | 12.12 (11.83, 12.40) | 125 | 32 (31.5) |
| Level of guardian education | ||||
| <High school | 294 | 11.96 (11.77, 12.16) | 311 | 88 (31.8) |
| High school or equivalent | 198 | 12.23 (11.99, 12.47) | 215 | 61 (23.3) |
| >High school | 390 | 12.19 (12.05, 12.33) | 415 | 132 (27.4) |
| Missing | 46 | 46 | ||
| BMI age- and sex-specific percentile# | ||||
| <50th | 217 | 12.52 (12.35, 12.69) | 238 | 47 (16.5) |
| 50–85th | 330 | 12.04 (11.91, 12.17) | 349 | 113 (29.1) |
| ≥85th | 381 | 11.97 (11.75, 12.19) | 400 | 141 (35.2) |
Abbreviations: BPA, bisphenol A; BMI, body mass index (kg/m).
Excluding girls who had not experienced menarche (n=59).
Percentages are weighted using population weights for the BPA subsample.
P <0.05 for difference in percentage of menarche below 12 years of age.
Table 3 presents odds ratios and 95% confidence intervals for the association between urinary BPA and age at menarche. In the model adjusted for creatinine, age, and NHANES cycle, adolescent girls with moderate BPA levels were 45 percent less likely to have early onset of menarche compared to those with the lowest levels (OR=0.55; 95% CI=0.31, 0.99). This association was no longer statistically significant after adjusting for race/ethnicity, parental education, country of birth, and BMI (OR=0.57; 95% CI=0.30, 1.08). In adjusted models, a 1-unit (ng/mL) increase in log-transformed continuous BPA was associated with a 15% decrease in the relative odds of early onset of menarche (OR=0.85; 95% CI=0.61, 1.18). Body size was significantly associated with early onset of menarche in adjusted analyses. Overweight girls were 52 percent more likely to have early onset of menarche compared to normal weight girls (OR=1.52; 95% CI=1.07, 2.16).
Table 3.
Association of urinary BPA concentrations and age at menarche <12 years, NHANES, 2003–2010
| Menarche <12/≥12 | Model 1a | Model 2b | |
|---|---|---|---|
|
| |||
| Urinary BPA concentrations (ng/mL) | OR (95% CI) | OR (95% CI) | |
|
|
|||
| Low (<1.9) | 111/225 | 1.00 (ref) | 1.00 (ref) |
| Moderate (1.9–4.1) | 97/227 | 0.55 (0.31, 0.99) | 0.57 (0.30, 1.08) |
| High (>4.1) | 93/234 | 0.73 (0.39, 1.37) | 0.77 (0.38, 1.55) |
| Log urinary BPA (ng/mL) | 301/686 | 0.84 (0.62, 1.15) | 0.85 (0.61, 1.18) |
| p for trend | 0.27 | 0.32 | |
| Body size | |||
| Under/normal weight | 1.00 (ref) | 1.00 (ref) | |
| Overweight/obese | 1.71 (1.20, 2.45) | 1.52 (1.07, 2.16) | |
Abbreviations: BPA, bisphenol A; OR, odds ratio; CI, confidence interval.
Model 1 is a logistic regression model adjusted for urinary creatinine (log-transformed mg/dL), age, and NHANES cycle.
Model 2 is a logistic regression model adjusted for race/ethnicity (non-Hispanic white, African American, Mexican American, other), parental education, country of birth, and body size in addition to model 1 covariates.
Overweight status additionally modified the association between BPA and age of menarche (p=0.08). Table 4 presents the odds ratios and 95% confidence intervals for the effect modification assessment with overweight status. Overweight girls in the lowest BPA tertile appear to be more likely to have early onset of menarche compared to normal weight girls in the lowest BPA tertile, though there is considerable variation in the estimate of the odds ratio (OR=1.65; 95% CI=0.84, 3.23).
Table 4.
Association of urinary BPA concentrations and age at menarche in relation to body size, NHANES, 2003–2010
| Not overweight | Overweight/obese | |||
|---|---|---|---|---|
|
|
|
|||
| Urinary BPA concentrations (ng/mL) | OR | 95% CI | OR | 95% CI |
| Low (<1.9) | 1.00 | (ref) | 1.65 | 0.84, 3.23 |
| Moderate (1.9–4.1) | 0.68 | 0.30, 1.53 | 0.74 | 0.41, 1.33 |
| High (>4.1) | 0.72 | 0.33, 1.58 | 1.36 | 0.60, 3.08 |
Abbreviations: BPA, bisphenol A; OR, odds ratio; CI, confidence interval.
ORs computed from model with interaction term (tertiles of BPA times body size), adjusted for age, urinary creatinine, race/ethnicity, parental education, NHANES cycle, and country of birth. p-interaction = 0.08
4. Discussion
We examined the association between exposure to BPA, a known EDC, and age of menarche, an endpoint in pubertal maturation. In this cross-sectional analysis of pooled NHANES 2003–2010 data, urinary BPA appears to be associated with overall delayed menarche, particularly for moderate levels of BPA exposure. However, while the magnitude of association remained virtually unchanged when likely confounders were considered, the result was no longer statistically significant. Further, the BPA-menarche association varied by levels of body size with overweight girls showing earlier development on average.
Epidemiologic studies have suggested that exposure to EDCs may be associated with earlier breast development and age of onset of menarche, though much of the literature is inconsistent (Ozen and Darcan, 2011). More specifically, studies in humans have shown exposure to dioxin (Warner et al., 2004) and PBBs (Blanck et al., 2000) to be associated with earlier onset of menarche. In a recent study using NHANES data, researchers found that high polybrominated diphenyl ether (PBDE) concentrations were associated with earlier age of menarche (Chen et al., 2011). However, other studies have shown delayed pubertal maturation with exposure to these chemicals. Particularly, one study showed that exposure to polychlorinated biphenyls (PCBs) may delay breast development (Wolff et al., 2008). In another study, researchers found exposure to dioxin to be associated with delayed breast development (Den Hond et al., 2002).
No significant associations in human studies have been found with exposure to BPA and pubertal outcomes (Wolff et al., 2008; Wolff et al., 2010). A recent study assessed the association between EDCs, including BPA, and age of menarche using NHANES 2003–2008 (Buttke et al., 2012). This study found an inverse relationship between age of menarche and urinary phenols. When the authors assessed BPA specifically, however, they found no significant associations. Differences between this particular study and the current analysis include sample size (due to the addition of the NHANES 2009–2010 cycle in the current analysis), as well as different statistical techniques.
The findings of the association of BMI with earlier menarche are consistent with previous studies. Research has shown that those girls with a higher BMI are at an increased risk for earlier pubertal maturation (Biro et al., 2006). However, our study showed the association between urinary BPA and age at menarche may be even stronger among overweight/obese girls with lower levels of BPA. We evaluated potential effect modification of the BPA-menarche association by overweight status. The results from the modification analysis showed that the BPA-menarche association differed by body size. Since BPA has been shown to be an obesogen (Wang et al., 2012), it is hypothesized that exposure to BPA could lead to a higher BMI in some girls, which could contribute to early menarche. Consequently, a recent study in Shanghai found BPA to be associated with more than a 2-fold risk of obesity in girls aged 9–12 (Li et al., 2013). Consequently, our model adjusting for body size may have been over-adjusted if body size truly mediates the effect of BPA. Obesogens are chemicals that can interfere with lipid metabolism and can potentially lead to obesity. The obesogen hypothesis supports that early life exposures to endocrine disrupting chemicals can predispose individuals to obesity (Janesick and Blumberg, 2011).
There are a few potential limitations in this analysis. The cross-sectional nature of this design limits the potential to examine causal associations. Further, a single, one-spot urine sample was collected for urinary BPA exposure. BPA is believed to have a short half-life in the human body, most of which is eliminated within 24 hours of exposure (Volkel et al., 2002). However, studies have shown that single spot urine samples of BPA may be representative of long-term averages and that spot samples can be used to characterize population distribution of intakes (Christensen et al., 2012; Teitelbaum et al., 2008; Vandenberg et al., 2007). It is not likely that this limitation contributed to our observed findings.
It is possible that there are other markers of pubertal maturation that may be related to BPA exposure. Breast development and presence of pubic hair are known to be the first markers of pubertal maturation. Pubertal maturation ends with the onset of menarche, which occurs approximately two years after breast development. In addition, earlier breast development may not lead to an earlier age of menarche (Biro et al., 2013). Thus, assessing onset of menarche does not allow one to examine changes in breast and pubic hair development. Earlier markers of pubertal maturation, such as thelarche and pubarche, may be more closely related with exposure to BPA. Additionally, age of menarche was assessed by recall, which may have introduced bias. Recall bias of age of onset of menarche may have been present, yet unlikely due to the short interval in recall (Koo and Rohan, 1997). There may have also been uncontrolled or residual confounding of other factors associated with the BPA exposure and menarche. We used post-pubertal BMI as a proxy for pre-pubertal BMI in our analysis. Although there is potential that some girls may have changed their overweight status based on categories of BMI, we do not believe this introduced significant bias. Finally, the current analysis was unable to adjust for genetic components of menarche, or mother’s age of menarche.
5. Conclusions
The current study takes advantage of a rich population based dataset with national biomonitoring data to examine the hypothesis that BPA, measured as urinary BPA, is associated with early onset of menarche in adolescent girls and the association is modulated by obesity status, measured as increased BMI. This study found urinary BPA was associated with overall delayed menarche in this sample of 987 adolescent girls from NHANES 2003–2010. Although a non-significant trend suggests increasing urinary BPA may be associated with delayed menarche in adolescent girls, these results are based on cross-sectional data. Results should be verified in carefully designed longitudinal cohort studies.
Highlights.
Bisphenol A (BPA) may be associated with putative markers of breast cancer risk.
Girls with moderate BPA levels were less likely to have early onset of menarche.
Body mass index appeared to modify the BPA-menarche association.
Results should be clarified in carefully designed longitudinal cohort studies.
Acknowledgments
This paper was supported in part by a training grant from the National Institutes of Health (T32ES007018).
Footnotes
Competing Financial Interests
All authors declare they have no actual or potential competing financial interests.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- Ashby J, Tinwell H. Uterotrophic activity of bisphenol A in the immature rat. Environ Health Perspect. 1998;106:719–20. doi: 10.1289/ehp.98106719. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Barr DB. Urinary creatinine concentrations in the U.S. population: implications for urinary biologic monitoring measurements. Environmental health perspectives. 2005;113:192–200. doi: 10.1289/ehp.7337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Betancourt AM. In utero exposure to bisphenol A shifts the window of susceptibility for mammary carcinogenesis in the rat. Environmental health perspectives. 2010;118:1614–9. doi: 10.1289/ehp.1002148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Biro F, Wolff M. Puberty as a Window of Susceptibility. In: Russo J, editor. Environment and Breast Cancer. Springer; New York: 2011. p. 31. [Google Scholar]
- Biro FM. Pubertal assessment method and baseline characteristics in a mixed longitudinal study of girls. Pediatrics. 2010;126:e583–90. doi: 10.1542/peds.2009-3079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Biro FM. Onset of breast development in a longitudinal cohort. Pediatrics. 2013;132:1019–27. doi: 10.1542/peds.2012-3773. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Biro FM. Influence of obesity on timing of puberty. International journal of andrology. 2006;29:272–7. doi: 10.1111/j.1365-2605.2005.00602.x. discussion 286–90. [DOI] [PubMed] [Google Scholar]
- Biro FM. Impact of yesterday’s genes and today’s diet and chemicals on tomorrow’s women. J Pediatr Adolesc Gynecol. 2009;22:3–6. doi: 10.1016/j.jpag.2008.12.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Black SR, Klein DN. Early menarcheal age and risk for later depressive symptomatology: the role of childhood depressive symptoms. J Youth Adolesc. 2012;41:1142–50. doi: 10.1007/s10964-012-9758-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Blanck HM. Age at menarche and tanner stage in girls exposed in utero and postnatally to polybrominated biphenyl. Epidemiology. 2000;11:641–7. doi: 10.1097/00001648-200011000-00005. [DOI] [PubMed] [Google Scholar]
- Buttke DE. Exposures to endocrine-disrupting chemicals and age of menarche in adolescent girls in NHANES (2003–2008) Environ Health Perspect. 2012;120:1613–8. doi: 10.1289/ehp.1104748. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Calafat AM. Exposure of the U.S. population to bisphenol A and 4-tertiary-octylphenol: 2003–2004. Environ Health Perspect. 2008;116:39–44. doi: 10.1289/ehp.10753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Centers for Disease Control and Prevention. Biomonitoring Summary: Bisphenol A. 2012. [Google Scholar]
- Chen A. Serum PBDEs and age at menarche in adolescent girls: analysis of the National Health and Nutrition Examination Survey 2003–2004. Environmental research. 2011;111:831–7. doi: 10.1016/j.envres.2011.05.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Christensen KL. Population variability of phthalate metabolites and bisphenol A concentrations in spot urine samples versus 24- or 48-h collections. Journal of exposure science & environmental epidemiology. 2012 doi: 10.1038/jes.2012.52. [DOI] [PubMed] [Google Scholar]
- Den Hond E. Sexual maturation in relation to polychlorinated aromatic hydrocarbons: Sharpe and Skakkebaek’s hypothesis revisited. Environmental health perspectives. 2002;110:771–6. doi: 10.1289/ehp.02110771. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ema M. Rat two-generation reproductive toxicity study of bisphenol A. Reproductive toxicology. 2001;15:505–23. doi: 10.1016/s0890-6238(01)00160-5. [DOI] [PubMed] [Google Scholar]
- Herman-Giddens ME. Recent data on pubertal milestones in United States children: the secular trend toward earlier development. International journal of andrology. 2006;29:241–6. doi: 10.1111/j.1365-2605.2005.00575.x. discussion 286–90. [DOI] [PubMed] [Google Scholar]
- Honma S. Low dose effect of in utero exposure to bisphenol A and diethylstilbestrol on female mouse reproduction. Reproductive toxicology. 2002;16:117–22. doi: 10.1016/s0890-6238(02)00006-0. [DOI] [PubMed] [Google Scholar]
- Janesick A, Blumberg B. Endocrine disrupting chemicals and the developmental programming of adipogenesis and obesity. Birth Defects Res C Embryo Today. 2011;93:34–50. doi: 10.1002/bdrc.20197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jenkins S. Oral exposure to bisphenol a increases dimethylbenzanthracene-induced mammary cancer in rats. Environ Health Perspect. 2009;117:910–5. doi: 10.1289/ehp.11751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kaplowitz PB. Link between body fat and the timing of puberty. Pediatrics. 2008;121(Suppl 3):S208–17. doi: 10.1542/peds.2007-1813F. [DOI] [PubMed] [Google Scholar]
- Kim K. Potencies of bisphenol A on the neuronal differentiation and hippocampal neurogenesis. J Toxicol Environ Health A. 2009;72:1343–51. doi: 10.1080/15287390903212501. [DOI] [PubMed] [Google Scholar]
- Koo MM, Rohan TE. Accuracy of short-term recall of age at menarche. Annals of human biology. 1997;24:61–4. doi: 10.1080/03014469700004782. [DOI] [PubMed] [Google Scholar]
- Li DK. Urine bisphenol-A level in relation to obesity and overweight in school-age children. PloS one. 2013;8:e65399. doi: 10.1371/journal.pone.0065399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Moral R. Effect of prenatal exposure to the endocrine disruptor bisphenol A on mammary gland morphology and gene expression signature. J Endocrinol. 2008;196:101–12. doi: 10.1677/JOE-07-0056. [DOI] [PubMed] [Google Scholar]
- Nagao T. Reproductive function in rats exposed neonatally to bisphenol A and estradiol benzoate. Reproductive toxicology. 1999;13:303–11. doi: 10.1016/s0890-6238(99)00017-9. [DOI] [PubMed] [Google Scholar]
- National Center for Health Statistics. NHANES Analytic and Reporting Guidlines. 2006. [Google Scholar]
- National Center for Health Statistics. 2003–2004 National Health and Nutrition Examination Survey: Laboratory procedures. 2010. [Google Scholar]
- Ozen S, Darcan S. Effects of environmental endocrine disruptors on pubertal development. Journal of clinical research in pediatric endocrinology. 2011;3:1–6. doi: 10.4274/jcrpe.v3i1.01. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Parent AS, et al. Early onset of puberty: tracking genetic and environmental factors. Hormone research. 2005;64(Suppl 2):41–7. doi: 10.1159/000087753. [DOI] [PubMed] [Google Scholar]
- Rasier G. Female sexual maturation and reproduction after prepubertal exposure to estrogens and endocrine disrupting chemicals: a review of rodent and human data. Mol Cell Endocrinol. 2006;254–255:187–201. doi: 10.1016/j.mce.2006.04.002. [DOI] [PubMed] [Google Scholar]
- Roy JR. Estrogen-like endocrine disrupting chemicals affecting puberty in humans--a review. Med Sci Monit. 2009;15:RA137–45. [PubMed] [Google Scholar]
- Teitelbaum SL. Temporal variability in urinary concentrations of phthalate metabolites, phytoestrogens and phenols among minority children in the United States. Environmental research. 2008;106:257–69. doi: 10.1016/j.envres.2007.09.010. [DOI] [PubMed] [Google Scholar]
- Tyl RW. Three-generation reproductive toxicity study of dietary bisphenol A in CD Sprague-Dawley rats. Toxicol Sci. 2002;68:121–46. doi: 10.1093/toxsci/68.1.121. [DOI] [PubMed] [Google Scholar]
- U.S. Environmental Protection Agency. Bisphenol A Action Plan. 2010. [Google Scholar]
- Vandenberg LN. Human exposure to bisphenol A (BPA) Reproductive toxicology. 2007;24:139–77. doi: 10.1016/j.reprotox.2007.07.010. [DOI] [PubMed] [Google Scholar]
- Volkel W. Metabolism and kinetics of bisphenol a in humans at low doses following oral administration. Chemical research in toxicology. 2002;15:1281–7. doi: 10.1021/tx025548t. [DOI] [PubMed] [Google Scholar]
- von Goetz N. Bisphenol a: how the most relevant exposure sources contribute to total consumer exposure. Risk Anal. 2010;30:473–87. doi: 10.1111/j.1539-6924.2009.01345.x. [DOI] [PubMed] [Google Scholar]
- Wang J. The environmental obesogen bisphenol A promotes adipogenesis by increasing the amount of 11beta-hydroxysteroid dehydrogenase type 1 in the adipose tissue of children. Int J Obes (Lond) 2012 doi: 10.1038/ijo.2012.173. [DOI] [PubMed] [Google Scholar]
- Warner M. Serum dioxin concentrations and age at menarche. Environmental health perspectives. 2004;112:1289–92. doi: 10.1289/ehp.7004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wolff MS. Environmental exposures and puberty in inner-city girls. Environmental research. 2008;107:393–400. doi: 10.1016/j.envres.2008.03.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wolff MS. Investigation of relationships between urinary biomarkers of phytoestrogens, phthalates, and phenols and pubertal stages in girls. Environmental health perspectives. 2010;118:1039–46. doi: 10.1289/ehp.0901690. [DOI] [PMC free article] [PubMed] [Google Scholar]