International Journal of Hygiene and Environmental Health 220 (2017) 43–50
Contents lists available at ScienceDirect
International Journal of Hygiene and Environmental Health journal homepage: www.elsevier.com/locate/ijheh
Urine bisphenol A and pubertal development in boys Ziliang Wang a,b , Dekun Li c , Maohua Miao b , Hong Liang b , Jianping Chen b , Zhijun Zhou d , Chunhua Wu d , Wei Yuan b,∗ a
School of Public Health, Fudan University, Shanghai, China Key Lab. of Reproduction Regulation of NPFPC, SIPPR, IRD, Fudan University, Shanghai, China Division of Research, Kaiser Foundation Research Institute, Kaiser Permanente, Oakland, California, USA d School of Public Health, Key Laboratory for Public Health Safety, Fudan University, Shanghai, China b c
a r t i c l e
i n f o
Article history: Received 30 May 2016 Received in revised form 8 October 2016 Accepted 10 October 2016 Keywords: Boy Bisphenol A Endocrine disruptor Puberty Tanner stage Testicular volume
a b s t r a c t Background: Bisphenol A (BPA) is an environmental endocrine disruptor and is found in many consumer products. Animal studies suggest that BPA may perturb pubertal development in males, although studies in humans are limited. Objective: This study investigated the association between BPA exposure and pubertal onset and progression among school-aged boys in Shanghai, China. Methods: A total of 671 boys aged 9–18 years from three schools (one elementary, one middle, and one high school) in Shanghai were enrolled in a cross-sectional study. Tanner stages for genital and pubic hair development and testicular volume were assessed by a specifically trained physician. Information concerning spermarche was self-reported. Urine samples were collected to examine peripubertal BPA exposure levels. Associations between BPA exposure and pubertal development, as indicated by the presence of different milestones in early puberty, mid-puberty and late puberty, were assessed using Poisson multivariate regression to derive adjusted prevalence ratios (PRs) and 95% confidence intervals (CIs). Results: Earlier onset of genital and pubic hair development was observed in boys with moderate BPA exposure compared with those exposed to the least BPA; the adjusted PRs were 1.31 (95%CI:1.03, 1.68) and 1.28 (95%CI:1.02, 1.60) for onset of genital maturation and pubic hair development, respectively. A similar trend was seen for onset of testicular development, although the association was not statistically significant. Conversely, compared with the lowest level of BPA exposure, moderate BPA exposure was associated with delayed presence of the late stage of genital development, with an adjusted PR of 0.78 (95%CI: 0.65, 0.92). A suggestive inverse association was also observed between BPA exposure and late progression of testicular development. Conclusions: Our findings indicate an association between peripubertal BPA exposure and earlier pubertal onset, but delayed pubertal progression, in boys. Longitudinal studies of male pubertal development with periodic follow-up are needed to verify these results. © 2016 Elsevier GmbH. All rights reserved.
1. Introduction Puberty is a period marked by rapid psychological, endocrine, and physical changes that bring about the transition from prepuberty to sexual maturity. Alterations in the timing of pubertal
Abbreviations: BMI, body mass index; BPA, bisphenol A; CI, confidence interval; EDC, endocrine-disrupting compound; HPLC, high-performance liquid chromatography; LOD, limit of detection; NOAEL, no observed adverse effect level; PR, prevalence ratio; TV, testicular volume. ∗ Corresponding author. E-mail address:
[email protected] (W. Yuan). http://dx.doi.org/10.1016/j.ijheh.2016.10.004 1438-4639/© 2016 Elsevier GmbH. All rights reserved.
onset or pace of pubertal development may not only have adverse effects on physical and sexual maturation, but also lead to problems with social, cognitive and behavioural development and adult health(Mendle et al., 2010; Walvoord, 2010). For example, earlier puberty is associated with an increased risk of metabolic syndrome, obesity, and diabetes in later life, while delayed puberty decreases bone mineral density in adults, potentially leading to a higher risk of fracture (Kindblom et al., 2006; Walvoord, 2010). Over the past 50 years, suggestive evidence of earlier onset of breast development and falling age at menarche has been observed in girls; however, data in boys are limited (Euling et al., 2008a). Although the available evidence is insufficient to draw conclusions
44
Z. Wang et al. / International Journal of Hygiene and Environmental Health 220 (2017) 43–50
regarding alterations in pubertal timing in boys, some studies with large populations and/or reliable puberty markers have reported an earlier age of pubertal onset in boys (Euling et al., 2008a; Sorensen et al., 2010). Changes in diet and activity are likely contributors to the observed alteration; however, environmental exposures may also play a role (Euling et al., 2008b). Endocrine-disrupting compounds (EDCs), as hormonally active chemicals, are of particular concern. Notably, children are often exposed to higher levels of EDCs and are more vulnerable to such environmental agents than adults (DiVall, 2013). Bisphenol A (BPA) is an EDC with both estrogenic and antiandrogenic effects. It is extensively used in the production of polycarbonate plastic and epoxy resins. Common products made with BPA include plastic containers, dental sealants and food and drink packaging materials, which we regularly come into contact with through daily life (Shelby, 2008). Studies in animals have yielded strong evidence of the association between BPA and adverse reproductive outcomes, including decreased uterine receptivity in females, and prostate pathogenesis and decreased sperm quality in males. Furthermore, in both animal and human studies, BPA is associated with reduced oocyte quality, altered steroidogenesis, and sexual dysfunction (Li et al., 2010; Peretz et al., 2014). However, there is insufficient evidence regarding the effects of BPA on pubertal development, particularly in boys. In animal studies, BPA exposure has been shown to disrupt sexual maturation, including vaginal opening and estrous cyclicity in females, and preputial separation, testicular descent, and steroidogenesis in males, although these results have been inconsistent (Kendig et al., 2012; Peretz et al., 2014; Tan et al., 2003). In a small number of epidemiological studies, associations between BPA exposure and delayed menarche in girls have been observed (Buttke et al., 2012; McGuinn et al., 2015). Data in boys are much more limited. One study reported that childhood BPA exposure was non-significantly associated with increased sex hormone-binding globulin levels and decreased total and free testosterone levels in boys, although no association with adrenarche or pubertal onset was found in this study (Ferguson et al., 2014). The present study was conducted among school-aged boys in Shanghai, China. The study objective was to investigate whether BPA exposure has any effect on pubertal development in boys. In addition to pubertal onset, which previous studies have frequently focused on (Zawatski and Lee, 2013), we examined the association between BPA and later pubertal stages to reflect the full picture of pubertal development associated with BPA exposure. 2. Materials and methods This was an ancillary study to a national survey of pubertal development and adolescent health in China, designed to collect anthropometric measures and related information in order to assess growth and pubertal development. The current study used additional questions concerning related factors of pubertal maturation and collected urine samples from boys in Jiading district, Shanghai, one of the sites of the national survey, from May 2011 to June 2011. A detailed description of the study has been presented previously (Li et al., 2013). The following sections focus on methodological issues particularly relevant to the present study. 2.1. Study population Three large schools (one elementary, one middle, and one high school) in Jiading district were selected. All students aged 9–18 years in grades 4 through 12 were considered eligible for inclusion in the study. Four classes of students from each grade were ran-
domly selected, recruiting approximately 80 boys from each grade (class size was typically around 40 students). Among 708 eligible boys, eight (1.1%) refused to participate. A further 18 boys did not provide urine samples and samples from 11 boys were accidentally damaged during transportation. Thus, 671 boys were included in final analyses, constituting 94.8% of the initial eligible population. The study was approved by the committees for protection of human subjects at Shanghai Institute of Planned Parenthood Research and School of Public Health, Fudan University. The parents of all the boys were sent a consent form with a detailed description of the study prior to enrolment. Parents were asked to inform teachers if they did not want their children to participate in the study. All boys were also informed in advance by their teachers of the study purpose and process, and the voluntary nature of participation; they were reminded again at the time of data collection. 2.2. Growth and pubertal assessment For each eligible participant, a physician conducted standardised physical examinations without knowledge of the boy’s BPA exposure status. Height and weight were measured whilst barefoot and clad only in light underclothes, in line with recommendations from the National Health and Nutrition Examination Survey (NHANES, 2007). Body mass index (BMI) was calculated as weight/height2 (kilograms per square meter). In addition, data from similar study populations (Jiang et al., 2004, 2007) were used to construct age- and gender- specific height and BMI distributions, allowing us to categorised height into percentile levels of <50th and ≥50th, while BMIs were categorised as normal weight, overweight or obese. Genital maturation and pubic hair development were graded from 1 (no development) to 5 (fully mature) by visual inspection according to established criteria (genital stages G1-G5 and pubic hair stages PH1-PH5) (Tanner and Whitehouse, 1976). Testicular volume (TV) was measured using an orchidometer. Genital stage 2 (G2) or TV > 3 mL for either testis was set as the benchmark for pubertal onset, and pubic hair stage 2 (PH2) was considered as another milestone in early puberty. Stage 5 for genitalia (G5), stage 5 for pubic hair (PH5) and TV > 20 mL were defined as milestones in late puberty (Biro et al., 1995; Bordini and Rosenfield, 2011). 2.3. In-person information collection At enrolment, each boy completed a questionnaire detailing a) age, dietary patterns, physical activity, sleep quality, health status, pubertal development; b) family income and residential history; and c) parental characteristics, including education and smoking behaviour. A food frequency questionnaire previously validated in a similar Chinese population (Hsu-Hage and Wahlqvist, 1992) was utilised to ascertain the boys’ dietary patterns. The published Children’s Depression Inventory (Kovacs, 1992) was used to assess depression status. For spermarche, which was considered a milestone occurring during mid-puberty (Bordini and Rosenfield, 2011; Zawatski and Lee, 2013), data were collected by asking the participant whether or not he had experienced ejaculations. 2.4. BPA exposure assessment One spot urine sample was collected from each participant. For each sample, the total urine BPA concentration (free plus conjugated species) was measured using modified high-performance liquid chromatography (HPLC), as described in previous studies (He et al., 2009). Analysis was conducted at the Department of Occupational Health and Toxicology, School of Public Health & WHO Collaborating Center for Occupational Health, Fudan University,
Z. Wang et al. / International Journal of Hygiene and Environmental Health 220 (2017) 43–50
Shanghai, China. Laboratory techniques and quality control protocols have been reported previously (He et al., 2009). The limit of detection (LOD) for BPA was 0.31 g/L, which is comparable to the published reports, including two Chinese studies (Calafat et al., 2008; Engel et al., 2014; He et al., 2009). Urinary creatinine was used to adjust for urine dilution.
2.5. Statistical analyses Since pubertal transition is a complex and integrated series of biological events, we examined associations between BPA level and different milestones (present or absent) representing onset of pubertal development (TV > 3 mL; G2; PH2), mid-puberty (spermarche) and late puberty (TV > 20 mL; G5; PH5). Modified Poisson regression with robust error variance was conducted to estimate prevalence ratios (PRs) rather than odds ratios as the outcomes (pubertal milestones) were not rare (Zou, 2004). PRs greater than 1 indicated that the exposed group was more likely to have reached the examined pubertal milestone, compared with the reference one. PRs and 95% confidence intervals (CIs) for onset of testicular development (TV > 3 mL vs. TV ≤ 3 mL), genital maturation [genital stage 2 or higher (G2 + ) vs. G1] and pubic hair development [pubic hair stage 2 or higher (PH2 + ) vs. PH1] were estimated to obtain the relative risk of reaching pubertal onset associated with BPA exposure. Similarly, PRs and 95% CIs for spermarche (present vs. absent) were examined in order to determine the presence of midpuberty milestone in relation to BPA. Additionally, considering that pubertal development follows a nonlinear trajectory and the association between pubertal onset and tempo remains unascertained (Marceau et al., 2011), we examined the late stages of testicular development (TV > 20 mL vs. TV ≤ 20 mL), genital maturation (G5 vs. G1-G4) and pubic hair development (PH5 vs. PH1-PH4) in relation to urine BPA to investigate the effects of BPA exposure on milestones in late puberty. We categorised the subjects into three groups using the LOD and the 75th percentile of creatininecorrected BPA concentrations (micrograms per gram creatinine) as cutoff points. In addition, values below the LOD were replaced √ with LOD/ 2. All creatinine-adjusted BPA concentrations were logtransformed to create a continuous BPA variable. We characterised creatinine-adjusted urine BPA concentrations as categorised levels or continuous log-transformed values in our statistical models. We identified covariates as factors potentially related to pubertal development and/or urine BPA concentration for the multivariate model. These factors included age at enrolment, height, BMI, household income, residence, parental education and smoking, lifestyle, depression status, food and nutrient intake. The model was developed by first assessing the univariate relation of each covariate to each pubertal development measure and retaining those with a p-value < 0.20. Backward selections were then used to iteratively exclude the least important covariates (retain p < 0.15). Covariates were included in the final model if they were retained for at least one pubertal development measure by the above criteria. In addition, we assumed that BMI might be in the causal pathway of the effect of BPA on pubertal development, and thus inclusion in the model could have resulted in over-adjustment. On the other hand, exogenous exposures are likely to operate jointly with BMI, which is a strong endogenous hormonal risk factor for pubertal development (Kaplowitz, 2008). Therefore, we first compared models with and without BMI adjustment and then investigated BMI as an effect modifier by adding an interaction term (BPA levels × BMI categorised by the overweight cutoff point) to examine the hypotheses. All statistical analyses were performed using Stata 12.0 (Stata Corp., LP, College Station, TX). Statistical significance was defined as p < 0.05.
45
3. Results 3.1. Demographic characteristics and BPA exposure Participants included in analyses (n = 671) had a mean (±SD) age of 13.43 ± 2.56 years. BPA was detected (concentrations > LOD) in 62.7% of urine samples (Table 1). The median (quartile range) urine BPA level was 2.06 (LOD-7.94) g/L. Higher BPA concentrations were found in urine from younger and obese boys, and the difference across age groups was significant (Table 1). Other characteristics including household income, residence, parental education and smoking behaviour, lifestyle, depression status and food and nutrient intake, were not associated with urine BPA concentrations. 3.2. Pubertal development Fig. 1 shows the prevalence of pubertal events representing early, mid- and late puberty at different ages. Almost all 13-yearold boys had begun pubertal development; the percentages were 100%, 94%, and 77.6% for boys reaching the milestones of TV > 3 mL, G2+ and PH2+, respectively. Of 16-year-old boys, 60.9% had reached a late stage of testicular development with TV > 20 mL, while 89.9% had reached G5. Nearly half of boys (49.3%) aged 16 years had reached PH5. Proportions of boys at different stages of testicular development (TV), genital maturation (G1-G5) and pubic hair development (PH1-PH5) revealed a convincing chronological order, changing reasonably in line with age (data not shown). However, only 37.3% of boys aged 14 years reported spermarche; this age has commonly been presented as the median age for spermarche (Ma et al., 2011; Sun et al., 2012). 3.3. BPA exposure and pubertal onset and progression Prevalence ratios and 95% CIs for any testicular development vs. no development in relation to urine BPA were estimated among boys younger than 13 years, as all participants beyond this age had already experienced onset of testicular development (Fig. 1). Similarly, onset of genital maturation and pubic hair development in relation to urine BPA were examined among boys younger than 14 and 15 years, respectively. BPA exposure was associated with earlier onset of genital maturation (G2 + ) and pubic hair development (PH2 + ). Adjusted PRs for reaching G2+ and PH2+ were 1.31 (95%CI:1.03, 1.68) and 1.28 (95%CI:1.02, 1.60) respectively for boys with moderate levels of urine BPA compared with those with the lowest levels. No significant association was observed with high BPA exposure (Table 2). A similar trend was seen for testicular development (TV > 3 mL), although the adjusted PR was not significant. Urine BPA level was not significantly associated with presence of spermarche, an indicator of mid-puberty (Table 3). However, inverse associations were observed on investigation of late pubertal stage and urine BPA (Table 4). Late stage of testicular development in relation to urine BPA was examined among boys aged 11 years or older, since no younger boys had reached this milestone (Fig. 1). Likewise, PRs and 95% CIs for late stage of genital maturation and pubic hair development in relation to urine BPA were estimated among boys aged 12 years or older. With regard to genital development, boys with moderate urine BPA levels were less likely to have reached G5 compared with those with the lowest BPA levels (PR = 0.78; 95%CI: 0.65, 0.92). A suggestive inverse association between urine BPA and presence of late testicular stage (by TV > 20 mL) was also seen with an adjusted PR of 0.80 (95%CI:0.62, 1.02; p = 0.08) for moderate BPA levels compared with the lowest levels. No significant association was detected between BPA level and presence of late pubic hair stage (by PH5). The results of mod-
46
Z. Wang et al. / International Journal of Hygiene and Environmental Health 220 (2017) 43–50
Table 1 Participant characteristics by urine BPAa level. Characteristic
N
Median BPAb (25th, 75th)
BPA level LOD-75th (%)
≥75th (%)
671
1.42 (0.20, 5.14)
37.3
37.8
24.9
Age (years)* <12 12–14 ≥15
232 231 208
1.90 (0.24, 6.07) 1.53 (0.26, 4.04) 0.86 (0.15, 4.67)
40.5 31.6 39.9
27.2 48.5 38.0
32.3 19.9 22.1
Heightc <50th ≥50th
213 458
2.06 (0.23, 5.87) 1.30 (0.20, 4.81)
35.2 38.2
36.2 38.7
28.6 23.1
BMIc Normal weight Overweight Obese
451 158 62
1.32 (0.19, 5.04) 1.70 (0.24, 5.11) 1.71 (0.23, 5.45)
38.8 36.0 29.0
36.6 39.2 43.6
24.6 24.7 27.4
Household characteristics Residence Rural Urban
107 514
1.78 (0.19, 5.14) 1.32 (0.20, 5.23)
33.6 38.3
42.1 36.4
24.3 25.3
Household income ≤Middle >Middle
423 239
1.42 (0.20, 5.12) 1.41 (0.21, 5.29)
36.2 39.3
39.2 35.2
24.6 25.5
Parental education ≤High School College & above
356 273
1.31 (0.20, 5.13) 1.80 (0.21, 5.28)
40.5 33.3
34.8 40.3
24.7 26.4
Parental smoking Yes No
401 228
1.43 (0.20, 5.01) 1.30 (0.21, 5.22)
37.7 37.7
38.2 37.3
24.2 25.0
Life style and mental health Unbalanced diet Yes No
260 359
1.16 (0.19, 4.86) 1.41 (0.21, 5.23)
40.4 36.8
36.9 38.2
22.7 25.1
Sports activity(30 min/day) Yes No
336 322
1.37 (0.20, 5.06) 1.45 (0.21, 5.44)
39.0 35.1
36.6 39.4
24.4 25.5
Sleeping quality Under/normal quality Good quality
224 397
0.97 (0.17, 3.71) 1.71 (0.22, 5.29)
40.6 36.0
39.3 37.8
20.1 26.2
Depression scores
262 270
1.60 (0.22, 4.73) 1.34 (0.19, 5.56)
34.0 37.4
43.5 35.2
22.5 27.4
Food and nutrient intake Fruits & Vegetables Everyday Not everyday
462 207
1.43 (0.20, 5.15) 1.41 (0.20, 5.04)
36.8 38.2
38.1 37.7
25.1 24.2
Soy-based foods Everyday Not everyday
174 493
2.03 (0.25, 5.48) 1.31 (0.20, 4.95)
33.9 38.3
38.5 37.9
27.6 23.7
Dairy Products Everyday Not everyday
337 332
1.74 (0.20, 5.57) 1.29 (0.21, 4.54)
35.6 39.2
37.1 38.6
27.3 22.3
Cereals Regularly(≥5days/week) Not regularly
193 476
1.11 (0.22, 4.81) 1.56 (0.20, 5.15)
38.3 36.8
37.3 38.2
24.4 25.0
Fish Regularly(≥5days/week) Not regularly
189 479
1.31 (0.23, 5.10) 1.60 (0.19, 5.28)
37.0 37.4
39.2 37.4
23.8 25.3
Meat Regularly(≥5days/week) Not regularly
86 585
1.57 (0.25, 5.40) 1.42 (0.20, 5.04)
38.4 37.1
33.7 38.5
27.9 24.4
Junk foods Regularly(≥5days/week) Not regularly
174 493
1.26 (0.20, 4.73) 1.58 (0.20, 5.40)
42.0 35.3
36.2 38.5
21.8 26.2
Total Growth measures
* a b c
p < 0.05. Creatinine-adjusted BPA (g/g creatinine). √ BPA concentrations below LOD were replaced with LOD/ 2. Height and BMI were categorized according to age- and gender-specific height and weight distribution produced using published data from similar study populations.
els using continuous log-transformed values were consistent with those using categorised levels, although the tests for linear trends did not show statistical significance (data not shown).
Overall, the subjects in the varied analyses mentioned above shared distributions similar to those of the total study population regarding the characteristics described in Table 1 (data not shown).
Z. Wang et al. / International Journal of Hygiene and Environmental Health 220 (2017) 43–50
E
A
79!PO
B
*
$JH
3HUFHQW
3+
$JH
D
3HUFHQW
79!PO
$JH
*
$JH
G
F
C
3HUFHQW
$JH
3HUFHQW
3HUFHQW
3HUFHQW
3HUFHQW
47
3+
$JH
6SHUPDUFKH
$JH
Fig. 1. Prevalence of pubertal events occurring in early, mid- and late puberty at different ages. (A) Prevalence of testicular volume greater than 3 mL; (B) prevalence of genital stage 2 or higher; (C) prevalence of pubic hair stage 2 or higher; (D) prevalence of spermarche; (E) prevalence of testicular volume greater than 20 mL; (F) prevalence of genital stage 5; (G) prevalence of pubic hair stage 5.
Removal of BMI at enrolment from the model did not materially change the findings (Tables 2–4). In addition, no statistically significant interaction between BPA and BMI was found in the effect modification assessment (data not shown).
4. Discussion The present cross-sectional study examined the relationship between exposure to BPA and pubertal development, including
48
Z. Wang et al. / International Journal of Hygiene and Environmental Health 220 (2017) 43–50
Table 2 Urine BPAa level in relation to pubertal onsetb . Reaching pubertal onsetc
BPA level
n (%)
Adjusted PRd
Adjusted PRe
66 (50.0%) 71 (67.6%)
≥75th
29 (34.9%)
1 (ref) 1.14 (0.95, 1.36) 1.04 (0.77, 1.41)
1 (ref) 1.14 (0.95, 1.37) 1.04 (0.77, 1.41)
48 (33.1%) 82 (59.4%)
≥75th
41 (39.4%)
1 (ref) 1.31** (1.03, 1.68) 1.16 (0.85, 1.58)
1 (ref) 1.31** (1.02, 1.67) 1.13 (0.83, 1.53)
50 (29.9%) 93 (53.1%)
≥75th
37 (30.6%)
1 (ref) 1.28** (1.02, 1.60) 1.11 (0.86, 1.42)
1 (ref) 1.28** (1.02, 1.60) 1.11 (0.86, 1.43)
Testicular volume (TV) > 3 mL
Genital stage 2 or higher (G2+)
Pubic hair stage 2 or higher (PH2+)
**
p < 0.05. Creatinine-adjusted BPA (g/g creatinine). b PRs and 95% CIs for associations between urine BPA level and pubertal onset of testicular development (TV > 3 mL vs. TV ≤ 3 mL), genital maturation (G2+ vs. G1) and pubic hair development (PH2+ vs. PH1). c Analyses for onset of testicular development were performed among boys younger than 13 years, while those for onset of genital maturation and pubic hair development were conducted among boys younger than 14 and 15 years, respectively. d Final model: adjusted for age, height, BMI, household income, parental education, sports activity, and intake of fruits and vegetables, soy-based foods, dairy products, cereals, fish, meat, and junk foods. e BMI removed from final model. a
Table 3 Urine BPAa level in relation to self-reported spermarche (present vs. absent)b . BPA level
n (%)
Adjusted PRc
Adjusted PRd
66 (27.7%) 68 (27.8%)
≥75th
31 (19.4%)
1 (ref) 1.08 (0.84, 1.39) 1.08 (0.77, 1.52)
1 (ref) 1.06 (0.83, 1.37) 1.08 (0.77, 1.52)
a
Creatinine-adjusted BPA (g/g creatinine). PRs and 95% CIs for associations between urine BPA level and self-reported spermarche (present vs. absent). c Final model: adjusted for age, height, BMI, household income, parental education, sports activity, and intake of fruits and vegetables, soy-based foods, dairy products, cereals, fish, meat, and junk foods. d BMI removed from final model. b
pubertal onset and late pubertal progression, in adolescent boys. Compared with the lowest BPA levels, BPA exposure at moderate levels was significantly associated with a higher likelihood of experiencing genital stage G2+, but a lower likelihood of experiencing the later stage G5. Similar patterns were observed for associations between BPA and testicular and pubic hair development. To our knowledge, this is the first account of an association between peripubertal BPA exposure and earlier pubertal onset accompanied by delayed pubertal progression in boys. The observed association between BPA exposure and earlier pubertal onset is supported by results from an animal study, which reported that BPA exposure below the established ‘no observed adverse effect level’ (NOAEL) was associated with accelerated preputial separation, a rat correlate of male pubertal onset, and increased circulating testosterone levels (Kendig et al., 2012). In contrast, other animal studies have indicated that high dose BPA exposure (above environmentally relevant doses) caused delay in male pubertal onset (Tan et al., 2003; Tyl et al., 2002). The fact that EDCs show non-monotonic dose-response relationships (Zawatski and Lee, 2013) may be a potential explanation for these conflicting results. Our study also presents the nonlinear dose-response effects of BPA, with no significant association observed for high levels of exposure. In line with our findings, a previous study that
investigated age at menarche among adolescent girls found significant effects for moderate, but not high, urine BPA levels (McGuinn et al., 2015). To our knowledge, only one well-designed epidemiologic study has examined the relationship between BPA exposure and puberty in boys; this study reported no significant association between childhood BPA exposure and adrenarche or pubertal onset (Ferguson et al., 2014). However, in this previous study, the nonlinear dose-response effects of BPA were not revealed as BPA exposure was assessed by continuous log-biomarker values. Furthermore, the small sample size of 113 boys limited the power of the study to detect any potential association. In addition to pubertal onset, the present study examined endpoints for late pubertal progression. Our results show that moderately elevated urine BPA levels are associated with earlier pubertal onset and delayed progression to the late stages of puberty. This is supported by the observation that a compensatory delay occurs in the post-pubertal period among adolescents who show earlier accelerated maturation (Marti-Henneberg and Vizmanos, 1997; Pantsiotou et al., 2008). The present study has several strengths. We examined different milestones representing early puberty, mid-puberty and late puberty to reflect the full picture of pubertal development associated with BPA exposure. Puberty is considered a complex sequence of biological events and male puberty typically begins with testicular enlargement (>3 mL) and genital development (G2). Pubarche (PH2) characteristically follows as another milestone in early puberty. Spermarche occurs during mid-puberty, and TV > 20 mL and genital and pubic hair development stage 5 (G5 and PH5) can be considered milestones in late puberty (Biro et al., 1995; Bordini and Rosenfield, 2011). Notably, pubertal development is a nonlinear process and the association between pubertal timing and tempo remains uncertain (Marceau et al., 2011). The effects of EDCs may vary with regard to different pubertal stages (Den Hond et al., 2002; Korrick et al., 2011). Furthermore, the present study collected information on dietary patterns and nutrient intake, and related variables were included in the final models as confounders. Analysis of pubertal outcomes
Z. Wang et al. / International Journal of Hygiene and Environmental Health 220 (2017) 43–50
49
Table 4 Urine BPAa level in relation to late pubertal progressionb . Reaching later stagec
BPA level
n (%)
Adjusted PRd
Adjusted PRe
69 (37.3%) 64 (28.8%)
≥75th
34 (29.6%)
1 (ref) 0.80* (0.62, 1.02) 0.98 (0.70, 1.35)
1 (ref) 0.77** (0.60, 0.98) 0.98 (0.72, 1.35)
89 (57.1%) 80 (41.9%)
≥75th
47 (51.1%)
1 (ref) 0.78*** (0.65, 0.92) 0.97 (0.79, 1.19)
1 (ref) 0.78*** (0.66, 0.93) 0.97 (0.79, 1.19)
40 (25.8%) 49 (25.7%)
≥75th
21 (22.8%)
1 (ref) 0.92 (0.66, 1.27) 0.87 (0.56, 1.35)
1 (ref) 0.84 (0.60, 1.18) 0.86 (0.56, 1.32)
Testicular volume (TV) > 20 mL
Genital stage 5 (G5)
Pubic hair stage 5 (PH5)
*
p < 0.1. p < 0.05. *** p < 0.01. a Creatinine-adjusted BPA (g/g creatinine). b PRs and 95% CIs for associations between urine BPA level and late pubertal stage of testicular development (TV > 20 mL vs. TV ≤ 20 mL), genital maturation (G5 vs. G1-G4) and pubic hair development (PH5 vs. PH1-PH4). c Analyses for late stage of testicular development were performed among boys aged 11 years or older, while those for late stage of genital maturation and pubic hair development were conducted among boys aged 12 years or older. d Final model: adjusted for age, height, BMI, household income, parental education, sports activity, and intake of fruits and vegetables, soy-based foods, dairy products, cereals, fish, meat, and junk foods. e BMI removed from final model. **
in epidemiological studies is complicated by the confounding factors of over-nutrition and BMI and the difficulty in distinguishing between the effects of nutrition and EDCs (Zawatski and Lee, 2013), particularly because foods are a common source of BPA exposure. In the present study, consumption of fish and meat, potential sources of BPA exposure (Larsson et al., 2014; Mervish et al., 2014), were associated with pubertal development (data not shown). On the other hand, adjustment for BMI did not materially alter the findings. Although BMI has been reported as a modifier of hormonal exposures in several studies examining EDCs in relation to female puberty (McGuinn et al., 2015; Wolff et al., 2010), no significant effect modification of BMI was observed in our study. Unlike the relationship in girls, the impact of body fat on onset of male puberty is uncertain (Bordini and Rosenfield, 2011; Kaplowitz, 2008). It should be noted that the present study measured BMI concurrently and therefore does not necessarily reflect the effects of prepubertal BMI, which are likely more relevant to pubertal onset. Several potential issues should be considered when interpreting findings from the present study. Firstly, stronger associations between urine BPA and markers of earlier pubertal onset were found for genital and pubic hair staging than for testicular volume. While early testicular size is due to follicle-stimulating hormone-mediated growth of seminiferous tubules, penile and scrotal maturation (G2) and pubic hair development (PH2) are primarily influenced by androgens (Zawatski and Lee, 2013). In animal models, BPA can disrupt the hypothalamic-pituitary-gonadal (HPG) axis, perturb gonadotropin secretion, and alter androgen synthesis (Peretz et al., 2014; Ramos et al., 2003). Considering the differences in developmental mechanisms, it may be possible that BPA exposure had a greater effect on G2 and PH2 than testicular volume. Another limitation to the study was that although our results validated the pubertal stage data assessed by the physician, selfreported spermarche was likely to be under-reported. Participants were not aware of the specific hypothesis of the study, thus, it is unlikely that under-reporting of spermarche was associated with their urine BPA concentrations. Therefore, under-reporting could
have led to non-differential misclassification, resulting in attenuation of the observed association. Furthermore, a single spot urine sample was collected to indicate BPA exposure, which may represent only recent exposure due to the relatively short half-life of BPA. On the other hand, previous studies have shown that a single spot urine sample of BPA may be representative of the long-term average (Teitelbaum et al., 2008) and provide a reasonably good measure of true BPA exposure level when the sample size is sufficiently large (Ye et al., 2011). Another consideration is that the proportion of boys with detectable BPA (62.7%) in this study was relatively lower compared with other studies of children and adolescents in China (Wang et al., 2014, 2012, 2015; Zhang et al., 2015); in these previous studies the rates of detection ranged from 84.9% to 99.9%, with the LOD below 0.1 g/L. The LOD of 0.31 g/L in the present study may be a potential reason. Two other studies in Chinese populations with LOD of 0.31 g/L and 0.4 g/L reported detection rates of 57% and 65%, respectively (Engel et al., 2014; He et al., 2009), which were comparable to our study. In addition, the cross-sectional nature of the study design limits the ability to examine a causal relationship. However, on the basis of our results, along with supportive animal data (Kendig et al., 2012; Ramos et al., 2003; Tan et al., 2003; Tyl et al., 2002), it is reasonable to conclude that BPA exposure could potentially disrupt pubertal development. Finally, peripubertal period is likely not the only critical window of exposure to factors which alter pubertal development. Perinatal exposure can also contribute to later development (Schoeters et al., 2008). Therefore, a lack of information on BPA exposure during other critical windows makes it difficult to rule out the impact of BPA exposure at earlier stages of development on puberty. 5. Conclusions Our findings provide epidemiological evidence that BPA exposure is linked to pubertal onset and pace of pubertal development in males. Compared with the lowest levels, moderate urine BPA levels were associated with earlier pubertal onset accompanied
50
Z. Wang et al. / International Journal of Hygiene and Environmental Health 220 (2017) 43–50
by delayed pubertal progression, especially for genital maturation. Further studies with longitudinal follow-up for pubertal development are needed to verify these findings.
Acknowledgments This work was supported by National Natural Scientific Foundation of China (81501318, 81270760) and Science and Technology Commission of Shanghai Municipality (15ZR1435100).
References Biro, F.M., Lucky, A.W., Huster, G.A., Morrison, J.A., 1995. Pubertal staging in boys. J. Pediatr. Us 127, 100–102. Bordini, B., Rosenfield, R.L., 2011. Normal pubertal development: part II: clinical aspects of puberty. Pediatr. Rev./Am. Acad. Pediatr. 32, 281–292. Buttke, D.E., Sircar, K., Martin, C., 2012. Exposures to endocrine-disrupting chemicals and age of menarche in adolescent girls in NHANES (2003–2008). Environ. Health Perspect. 120, 1613–1618. Calafat, A.M., Ye, X., Wong, L.Y., Reidy, J.A., Needham, L.L., 2008. Exposure of the U.S. population to bisphenol A and 4-tertiary-octylphenol: 2003–2004. Environ. Health Perspect. 116, 39–44. Den Hond, E., Roels, H.A., Hoppenbrouwers, K., Nawrot, T., Thijs, L., Vandermeulen, C., Winneke, G., Vanderschueren, D., Staessen, J.A., 2002. Sexual maturation in relation to polychlorinated aromatic hydrocarbons: sharpe and Skakkebaek’s hypothesis revisited. Environ. Health Perspect. 110, 771–776. DiVall, S.A., 2013. The influence of endocrine disruptors on growth and development of children. Curr. Opin. Endocrinol. 20, 50–55. Engel, L.S., Buckley, J.P., Yang, G., Liao, L.M., Satagopan, J., Calafat, A.M., Matthews, C.E., Cai, Q., Ji, B.T., Cai, H., Engel, S.M., Wolff, M.S., Rothman, N., Zheng, W., Xiang, Y.B., Shu, X.O., Gao, Y.T., Chow, W.H., 2014. Predictors and variability of repeat measurements of urinary phenols and parabens in a cohort of Shanghai women and men. Environ. Health Perspect. 122, 733–740. Euling, S.Y., Herman-Giddens, M.E., Lee, P.A., Selevan, S.G., Juul, A., Sorensen, T.I.A., Dunkel, L., Himes, J.H., Teilmann, G., Swan, S.H., 2008a. Examination of US puberty-timing data from 1940 to 1994 for secular trends: panel findings. Pediatrics 121, S172–S191. Euling, S.Y., Selevan, S.G., Pescovitz, O.H., Skakkebaek, N.E., 2008b. Role of environmental factors in the timing of puberty. Pediatrics 121 (Suppl. 3), S167–171. Ferguson, K.K., Peterson, K.E., Lee, J.M., Mercado-Garcia, A., Blank-Goldenberg, C., Tellez-Rojo, M.M., Meeker, J.D., 2014. Prenatal and peripubertal phthalates and bisphenol A in relation to sex hormones and puberty in boys. Reprod. Toxicol. 47, 70–76. He, Y., Miao, M., Herrinton, L.J., Wu, C., Yuan, W., Zhou, Z., Li, D.K., 2009. Bisphenol A levels in blood and urine in a Chinese population and the personal factors affecting the levels. Environ. Res. 109, 629–633. Hsu-Hage, B.H., Wahlqvist, M.L., 1992. A food frequency questionnaire for use in Chinese populations and its validation. Asia Pac. J. Clin. Nutr. 1, 211–223. Jiang, Y.F., Cole, T., Pan, H.Q., Ju, M.F., Lin, Z.F., Dong, X.Y., Zhang, L., 2004. Construct body mass index centile curves and cut off points for overweight and obesity for Shanghai children. Chinese J. Child Health Care 12, 461–464. Jiang, Y.F., Cole, T., Pan, H.Q., Lin, Z.F., Ju, M.F., Zhang, L., 2007. Construction of height and weight percentile references for Shanghai children and adolescents. Shanghai J. Preventive Med. 19, 544–547. Kaplowitz, P.B., 2008. Link between body fat and the timing of puberty. Pediatrics 121, S208–S217. Kendig, E.L., Buesing, D.R., Christie, S.M., Cookman, C.J., Gear, R.B., Hugo, E.R., Kasper, S.N., Kendziorski, J.A., Ungi, K.R., Williams, K., Belcher, S.M., 2012. Estrogen-like disruptive effects of dietary exposure to bisphenol a or 17 alpha-ethinyl estradiol in CD1 mice. Int. J. Toxicol. 31, 537–550. Kindblom, J.M., Lorentzon, M., Norjavaara, E., Hellqvist, A., Nilsson, S., Mellstrom, D., Ohlsson, C., 2006. Pubertal timing predicts previous fractures and BMD in young adult men: the GOOD study. J. Bone Miner. Res. 21, 790–795. Korrick, S.A., Lee, M.M., Williams, P.L., Sergeyev, O., Burns, J.S., Patterson, D.G., Turner, W.E., Needham, L.L., Altshul, L., Revich, B., Hauser, R., 2011. Dioxin exposure and age of pubertal onset among russian boys. Environ. Health Perspect. 119, 1339–1344. Kovacs, M., 1992. Children Depression Inventory CDI: Manual. Multi-Health Systems, New York. Larsson, K., Ljung Bjorklund, K., Palm, B., Wennberg, M., Kaj, L., Lindh, C.H., Jonsson, B.A., Berglund, M., 2014. Exposure determinants of phthalates, parabens, bisphenol A and triclosan in Swedish mothers and their children. Environ. Int. 73, 323–333. Li, D., Zhou, Z., Qing, D., He, Y., Wu, T., Miao, M., Wang, J., Weng, X., Ferber, J.R., Herrinton, L.J., Zhu, Q., Gao, E., Checkoway, H., Yuan, W., 2010. Occupational exposure to bisphenol-A (BPA) and the risk of self-reported male sexual dysfunction. Hum. Reprod. (Oxford England) 25, 519–527. Li, D.K., Miao, M.H., Zhou, Z.H., Wu, C.H., Shi, H.J., Liu, X.Q., Wang, S.Q., Yuan, W., 2013. Urine bisphenol-A level in relation to obesity and overweight in school-Age children. PLoS One 8.
Ma, H.M., Chen, S.K., Chen, R.M., Zhu, C., Xiong, F., Li, T., Wang, W., Liu, G.L., Luo, X.P., Liu, L., Du, M.L., 2011. Pubertal development timing in urban Chinese boys. Int. J. Androl. 34, e435–445. Marceau, K., Ram, N., Houts, R.M., Grimm, K.J., Susman, E.J., 2011. Individual differences in boys’ and girls’ timing and tempo of puberty: modeling development with nonlinear growth models. Dev. Psychol. 47, 1389–1409. Marti-Henneberg, C., Vizmanos, B., 1997. The duration of puberty in girls is related to the timing of its onset. J. Pediatr. 131, 618–621. McGuinn, L.A., Ghazarian, A.A., Su, L.J., Ellison, G.L., 2015. Urinary bisphenol A and age at menarche among adolescent girls: evidence from NHANES 2003–2010. Environ. Res. 136, 381–386. Mendle, J., Harden, K.P., Brooks-Gunn, J., Graber, J.A., 2010. Development’s tortoise and hare: pubertal timing, pubertal tempo, and depressive symptoms in boys and girls. Dev. Psychol. 46, 1341–1353. Mervish, N., McGovern, K.J., Teitelbaum, S.L., Pinney, S.M., Windham, G.C., Biro, F.M., Kushi, L.H., Silva, M.J., Ye, X., Calafat, A.M., Wolff, M.S., 2014. Dietary predictors of urinary environmental biomarkers in young girls, BCERP, 2004–7. Environ. Res. 133, 12–19. NHANES, 2007. Anthropometry Procedures Manual. National Health and Nutrition Examination Survey. Pantsiotou, S., Papadimitriou, A., Douros, K., Priftis, K., Nicolaidou, P., Fretzayas, A., 2008. Maturational tempo differences in relation to the timing of the onset of puberty in girls. Acta Paediatr. 97, 217–220. Peretz, J., Vrooman, L., Ricke, W.A., Hunt, P.A., Ehrlich, S., Hauser, R., Padmanabhan, V., Taylor, H.S., Swan, S.H., VandeVoort, C.A., Flaws, J.A., 2014. Bisphenol a and reproductive health: update of experimental and human evidence, 2007–2013. Environ. Health Perspect. 122, 775–786. Ramos, J.G., Varayoud, J., Kass, L., Rodriguez, H., Costabel, L., Munoz-De-Toro, M., Luque, E.H., 2003. Bisphenol A induces both transient and permanent histofunctional alterations of the hypothalamic-pituitary-gonadal axis in prenatally exposed male rats. Endocrinology 144, 3206–3215. Schoeters, G., Den Hond, E., Dhooge, W., Van Larebeke, N., Leijs, M., 2008. Endocrine disruptors and abnormalities of pubertal development. Basic Clin. Pharmacol. 102, 168–175. Shelby, M.D., 2008. NTP-CERHR monograph on the potential human reproductive and developmental effects of bisphenol A. NTP CERHR Monogr v (vii–ix) (1–64 passim). Sorensen, K., Aksglaede, L., Petersen, J.H., Juul, A., 2010. Recent changes in pubertal timing in healthy Danish boys: associations with body mass index. J. Clin. Endocrinol. Metab. 95, 263–270. Sun, Y., Tao, F., Su, P.Y., 2012. National estimates of pubertal milestones among urban and rural Chinese boys. Ann. Hum. Biol. 39, 461–467. Tan, B.L.L., Kassim, N.M., Mohd, M.A., 2003. Assessment of pubertal development in juvenile male rats after sub-acute exposure to bisphenol A and nonylphenol. Toxicol. Lett. 143, 261–270. Tanner, J.M., Whitehouse, R.H., 1976. Clinical longitudinal standards for height, weight, height velocity, weight velocity, and stages of puberty. Arch. Dis. Child. 51, 170–179. Teitelbaum, S.L., Britton, J.A., Calafat, A.M., Ye, X., Silva, M.J., Reidy, J.A., Galvez, M.P., Brenner, B.L., Wolff, M.S., 2008. Temporal variability in urinary concentrations of phthalate metabolites, phytoestrogens and phenols among minority children in the United States. Environ. Res. 106, 257–269. Tyl, R.W., Myers, C.B., Marr, M.C., Thomas, B.F., Keimowitz, A.R., Brine, D.R., Veselica, M.M., Fail, P.A., Chang, T.Y., Seely, J.C., Joiner, R.L., Butala, J.H., Dimond, S.S., Cagen, S.Z., Shiotsuka, R.N., Stropp, G.D., Waechter, J.M., 2002. Three-generation reproductive toxicity study of dietary bisphenol A in CD Sprague-Dawley rats. Toxicol. Sci. 68, 121–146. Walvoord, E.C., 2010. The timing of puberty: is it changing? Does it matter. J. Adolesc. Health 47, 433–439. Wang, H.X., Zhou, Y., Tang, C.X., Wu, J.G., Chen, Y., Jiang, Q.W., 2012. Association between bisphenol A exposure and body mass index in Chinese school children: a cross-sectional study. Environ. Health: Global Access Sci. Source 11, 79. Wang, B., Wang, H., Zhou, W., He, Y., Zhou, Y., Chen, Y., Jiang, Q., 2014. Exposure to bisphenol A among school children in eastern China: a multicenter cross-sectional study. J. Exposure Sci. Environ. Epidemiology 24, 657–664. Wang, N., Zhou, Y., Fu, C., Wang, H., Huang, P., Wang, B., Su, M., Jiang, F., Fang, H., Zhao, Q., Chen, Y., Jiang, Q., 2015. Influence of bisphenol a on thyroid volume and structure independent of iodine in school children. PLoS One 10, e0141248. Wolff, M.S., Teitelbaum, S.L., Pinney, S.M., Windham, G., Liao, L., Biro, F., Kushi, L.H., Erdmann, C., Hiatt, R.A., Rybak, M.E., Calafat, A.M., Ctr, B.C., Ctr, E.R., 2010. Investigation of relationships between urinary biomarkers of phytoestrogens, phthalates, and phenols and pubertal stages in girls. Environ. Health Perspect. 118, 1039–1046. Ye, X., Wong, L.Y., Bishop, A.M., Calafat, A.M., 2011. Variability of urinary concentrations of bisphenol A in spot samples, first morning voids, and 24-hour collections. Environ. Health Perspect. 119, 983–988. Zawatski, W., Lee, M.M., 2013. Male pubertal development: are endocrine-disrupting compounds shifting the norms? J. Endocrinol. 218, R1–R12. Zhang, M., Duan, Z., Wu, Y., Liu, Z., Li, K., Wang, L., 2015. Occurrence and profiles of the artificial endocrine disruptor bisphenol a and natural endocrine disruptor phytoestrogens in urine from children in China. Int. J. Environ. Res. Public Health 12, 15110–15117. Zou, G.Y., 2004. A modified Poisson regression approach to prospective studies with binary data. Am. J. Epidemiol. 159, 702–706.