You are here

Article Text

Article menu

Download PDFPDF

Research report
Urinary bisphenol A concentrations are associated with abnormal liver function in the elderly: a repeated panel study
  1. Mee-Ri Lee1,
  2. Hyunseung Park1,
  3. Sanghyuk Bae1,
  4. Youn-Hee Lim2,
  5. Jin Hee Kim3,4,
  6. Soo-Hun Cho1,
  7. Yun-Chul Hong1,3,4
  1. 1Department of Preventive Medicine, Seoul National University College of Medicine, Seoul, Korea
  2. 2Department of Epidemiology and Biostatistics, School of Public Health, Seoul National University, Seoul, Korea
  3. 3Institute of Environmental Medicine, Seoul National University Medical Research Center, Seoul, Korea
  4. 4Environmental Health Center, Seoul National University College of Medicine, Seoul, Korea
  1. Correspondence to Dr Yun-Chul Hong, Department of Preventive Medicine, Seoul National University College of Medicine, 103 Daehakro, Jongrogu, Seoul 110-744, Republic of Korea; ychong1{at}snu.ac.kr

Abstract

Background Bisphenol A (4,40-isopropylidenediphenol, BPA) is known to adversely affect various organs. The liver is reported to be affected by BPA in animal studies. However, there are few studies in humans on the effects of BPA on the liver. Therefore, we evaluated the relationship between urinary BPA levels and liver function in elderly subjects using repeated measurements.

Methods From 2008 to 2010, a total of 560 elderly subjects residing in Seoul were each evaluated up to three times. At the first visit, demographic data, environmental exposure and lifestyle information were obtained from a systemised questionnaire. At each visit, blood and urine samples were collected and stored for analysis. Linear mixed and GLIMMIX model analyses were performed after adjusting for age, sex, Body Mass Index, alcohol consumption, urinary cotinine concentrations, exercise frequency, and low-density lipoprotein cholesterol level.

Results The mean urinary BPA concentration was 1.13 μg/g creatinine. Significant relationships were observed between urinary BPA and aspartate aminotransferase, alanine aminotransferase, and gamma-glutamyl transferase after adjusting for potential confounders (p<0.05). When subjects were grouped according to urinary BPA concentrations divided by the median value, higher urinary BPA concentrations were associated with increased abnormal liver function (OR 2.66; 95% CI 1.15 to 5.90).

Conclusions Community-level exposure to BPA was associated with abnormal liver function in the elderly, indicating that more stringent control of BPA is necessary to protect susceptible populations.

  • AGEING
  • ENVIRONMENTAL HEALTH
  • EPIDEMIOLOGY

https://doi.org/10.1136/jech-2013-202548

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Bisphenol A (4,40-isopropylidenediphenol, BPA) is one of the world's highest volume chemicals by mass, which increases by 6–10% annually.1 People are exposed to BPA mainly in polycarbonate plastics and epoxy resins; moreover, BPA residues are found in food and beverage packaging, drinking water, bathwater and baby products.2–4 BPA has been found in most urine samples tested during national surveys from the USA, Canada, and Germany.2 ,5 ,6 Besides urine, other human biological samples such as follicular fluid, maternal and fetal serum, and amniotic fluid also contain BPA.7 ,8

    The ubiquity of BPA exposure and its endocrine-disrupting potential raise concerns about its effects on health despite the molecule's short half-life.9 BPA mimics the action of natural oestrogen by binding to oestrogen receptors (ERα and ERβ) and is, thus, an endocrine-disrupting chemical.10 ,11 However, BPA binds to oestrogen receptors 1000–5000 times less potently than 17-β-estradiol.12 In rats, BPA treatment results in early sexual maturation,13 an elevated incidence of reproductive organ lesions,14 prostate developmental abnormalities,15 and decreased sperm production.4 However, recent animal and in vitro studies suggest that BPA exerts additional health effects on other systems, such as central nervous system activity,16 thyroid hormone and pancreatic β-cell function.17 ,18

    Another rarely studied topic is the effect of BPA exposure on liver function. A study on BPA-treated rats showed elevated aspartate aminotransferase (AST), alanine aminotransferase (ALT), and lactate dehydrogenase (LDH) levels as well as marked defects in liver morphology.19 A recent cross-sectional human study of 1455 Americans demonstrates a statistically significant association between higher BPA concentrations and abnormal concentrations of three liver enzymes, that is, γ-glutamyl transferase (γ-GTP), alkaline phosphatase (ALP) and LDH.20

    However, the relationship between BPA exposure and liver function abnormality remains unclear, although some studies suggest possible mechanisms for liver damage, such as oxidative stress19 ,21 and direct apoptotic activities.21 ,22

    Since current information about the effect of BPA on liver function in humans is very limited despite suggestive findings in animal studies, the present panel study investigated the associations between BPA exposure and liver function. To this end, we examined whether urinary BPA concentrations are associated with liver enzyme concentrations using repeated measurements. We chose AST, ALT and γ-GTP to represent liver function, as they are sensitive markers for representing hepatocellular damage.

    Methods

    Study populations

    A panel study, which is a longitudinal study in which variables are measured repeatedly, was performed at a community welfare centre in the northern area of Seoul between 2008 and 2010. We recruited subjects who volunteered to participate in the study. A total of 560 elderly people aged ≥60 years visited a community welfare centre as many as five times for medical examinations during the study period (twice in 2008, once in 2009 and twice in 2010). Urine samples were collected at each visit. However, since blood samples were obtained annually (ie, not obtained at the second or fifth visits), we obtained as many as three blood samples from each individual. Revisit intervals during follow-up were longer than 3 months. Urine samples (50 mL) were collected from each subject, divided into four 12 mL tubes and stored at −20°C in a freezer. Blood samples (16 mL) were collected from each subject and divided into a 7 mL serum separation tube, 3 mL EDTA tube, and 6 mL trace element tube. Serum was separated after centrifugation at 2500 rpm for 10 min and kept at −70°C in a deep freezer until analysis. Every recruited subject was given a structured questionnaire including questions about demographics (ie, age, sex, education, marriage and income level), lifestyle habits (ie, smoking, alcohol consumption and exercise) and medical history. The Institutional Review Board of the Seoul National University Hospital approved the study protocol. All subjects provided written informed consent prior to enrolment.

    Urinary bisphenol A measurement

    Spot urine samples (50 mL) were collected in conical tubes (SPL Lifesciences, Pocheon, Gyunggi-do, Korea) from 10:00 to 12:00, stored at −20°C in a freezer, and sent to NeoDin Medical Institute (Seoul, Korea) for analysis within 90 min. Blank, standard solution, quality control materials (48th G-EQUAS A & B), and urine sample were placed into tubes in 500 mL aliquots; buffered with 30 μL 2 M sodium acetate (pH 5.0); and spiked with 25 μL ISTD BPA (RING-13C12, 99%, Cambridge Isotope Lab.) and 10 μL β-glucuronidase/sulfatase type HP-2S from Helix pomatia (Sigma, St Louis, Missouri, USA). The samples were incubated at 37°C for 3 h to deconjugate the glucuronidated BPA. Following incubation, the samples were cooled, 100 μL 2 N HCl and 4 mL ethyl acetate were added, and the samples were centrifuged at 2500 rpm for 5 min. The extract was dried with nitrogen gas and reconstituted with 1 mL high-performance liquid chromatography (HPLC)-grade H2O in a 2 mL glass vial. We subsequently performed liquid–liquid extraction (LLE). The analytic column was an Agilent Eclipse plus C18 (3.5 μm, 2.1×100 mm). The mobile phase was acetonitrile/water (60:40, v/v) at a flow rate of 0.4 mL/min. We measured urinary BPA concentrations using HPLC tandem mass spectrometry (HPLC: Agilent 1200, USA; MS/MS: Agilent 6410 Triple Quad LCMS, Agilent, USA). We used reagents with deionised water, acetonitrile, ethyl acetate, sodium acetate buffer and β-glucuronidase from Escherichia coli (SIGMA G0876; minimum, 85 000 units/mL). We measured total BPA, including free and conjugated BPA. The limit of detection (LOD) was calculated with a detection of signal-to-noise ratio of 3 at the lowest concentration; the LOD for urinary BPA was 0.005 μg/L. The quality control method was adopted from the Clinical and Laboratory Standards Institute (CLSI) guideline. Accuracy was 99.7%±0.038%, the coefficient of variation of precision was 1.0–4.7, the coefficient of variation of reproducibility was 0.5–5.3, and recovery was 100.1–104.4%.

    Creatinine measurement

    For creatinine measurement, CREA (Roche, Indianapolis, Indiana, USA) reagent was used with a Hitachi 7600 (Hitachi, Tokyo, Japan) with a kinetic colorimetric assay (rate blanked and compensated). We used value (μg/L)/creatinine (g/L) for dilution correction in the analyses.

    Liver enzymes measurement

    Blood samples (up to 3 mL) were collected from each participant using a BD vacutainer (Becton Dickinson, Franklin Lakes, New Jersey, USA) containing K2EDTA (Becton Dickinson) and preserved at −70°C. Serum AST, ALT and γ-GTP were subsequently analysed by an autobiochemical analyser (Hitachi 7600-II, Hitachi High-Technologies, Japan). Pureauto S AST, Pureauto S ALT, and Pureauto S γ-GTP (Daiichi Pure Chemicals, Tokyo, Japan) were used as reagents.

    Data analysis

    Among 560 subjects, 25 whose blood and urine samples were unavailable, and another 52 who visited the centre only once were excluded. Additionally, five subjects who reported currently having any type of viral hepatitis, fatty liver disease, liver cancer, or any other liver disease were excluded. Finally, 478 subjects and 1022 samples which were urine and blood sample pairs were included in the analysis. Creatinine-adjusted urinary BPA concentrations were used to control for different urinary excretion rates. Arithmetic means of urinary BPA concentrations were calculated according to demographic subgroups (ie, sex, age, smoking, alcohol consumption, physical exercise and Body Mass Index (BMI) groups). We used linear mixed-model analysis to evaluate the relationship between urinary BPA and the levels of three liver enzymes. Because the concentrations of the three liver enzymes were not normally distributed, log-transformed values were used in the analysis. Linear mixed models were adjusted for confounders considered to affect liver function, including age, sex, BMI, alcohol consumption, exercise, urinary cotinine levels and low-density lipoprotein cholesterol levels (LDL-C). The ORs of liver function abnormality associated with higher urinary BPA concentrations were calculated using the GLIMMIX model. To calculate ORs, urinary BPA and liver enzyme concentrations were transformed into binary variables. For urinary BPA, samples were divided by the median concentrations of men and women (0.57 and 0.67 μg/g creatinine, respectively). We set the cutoff values for elevated AST, ALT and γ-GTP at 35, 45 and 60 IU/L, respectively. Then, we set the criteria for clinically significant abnormality for liver function test as >1.5 times the cutoff value of any one of the liver enzymes. The GLIMMIX models were also adjusted by age, sex, BMI, alcohol consumption, exercise, urinary cotinine levels and LDL-C. BMI was calculated from measured height and weight at each visit. We used urinary cotinine levels to represent exposure to direct and indirect smoking; the detection range for cotinine was 1–10 000 µg/L. We imputed 0.5 μg/g creatinine for values lower than 1 μg/g creatinine, and 15 000 μg/g creatinine for values greater than 10 000 μg/g creatinine. LDL-C levels were calculated from measured values of total cholesterol, triglycerides and high-density lipoprotein cholesterol. Age, BMI, urinary cotinine levels and LDL-C levels were included in linear mixed and GLIMMIX models as continuous variables. We also assessed current alcohol consumption status (yes/no) and regular exercise status (yes/no) using structured questionnaires, and input them into the linear mixed and GLIMMIX models as categorical variables. Figures were drawn using generalised additive mixed models of R V.2.12.2 (The Comprehensive R Archive Network: http://cran.r-project.org). Linear mixed and GLIMMIX models were made using SAS V.9.2 (SAS Institute, Cary, North Carolina, USA). The level of significance was set at p<0.05.

    Results

    Table 1 shows creatinine-adjusted urinary BPA concentrations according to demographic characteristics of study subjects at the first visit. Participants were 123 (25.7%) men and 355 (74.3%) women. Most were aged between 60 and 79 years. Urinary BPA levels were significantly higher in women (1.28±1.95 μg/g creatinine) than in men (0.88±1.33 μg/g creatinine) (p=0.03).

    View this table:
    Table 1

    Demographic characteristics of the study population at the first visit

    Creatinine-adjusted urinary BPA concentrations were similar to the unadjusted concentrations (see online supplementary table S1). There were two extreme urinary BPA concentrations of 35.16 and 43.76 μg/g creatinine. The distributions of the levels of the three liver enzymes are shown in online supplementary figure S1. Over 90% of liver enzyme levels were within their normal range.

    The levels of all three enzymes were significantly elevated with increased urinary BPA concentrations in the model fully adjusted for age, sex, BMI, LDL-C levels, alcohol consumption, physical exercise and urinary cotinine levels (table 2). The results were similar among differently adjusted models. Moreover, the results did not differ when the same models were analysed after excluding the two extremely high BPA concentrations mentioned above (see online supplementary table S2).

    View this table:
    Table 2

    Linear mixed model analysis for liver enzymes and urinary BPA

    The non-parametric associations of urinary BPA concentrations with the levels of the three liver enzymes are shown in figure 1. The levels of all three liver enzymes showed near-linear associations with the BPA concentrations regardless of the statistical model.

    Figure 1
    Figure 1

    Generalised additive model for urinary bisphenol A (BPA) concentrations and three liver enzymes. *Additionally adjusted for age, sex, Body Mass Index (BMI), LDL-C levels, alcohol consumption, physical exercise, and urinary cotinine levels.

    The OR for abnormal liver function which was defined as AST >52.5 IU/L, ALT >67.5 IU/L, or γ-GTP >90 IU/L, for higher urinary BPA concentration, was 2.66 (95% CI 1.15 to 5.90) in the fully adjusted model (table 3). Excluding the two outliers did not alter the results (see online supplementary table S3).

    View this table:
    Table 3

    ORs of liver damage according to urinary BPA levels

    Supplementary tables S4 and S5 were created using the average urinary BPA of each subject, similar to tables 2 and 3. The results were significant for AST but not for ALT or γ-GTP (see online supplementary table S4). The results shown in supplementary table S5 are similar to those shown in table 3.

    Discussion

    In the present study, urinary BPA concentrations were associated with elevated liver enzyme concentrations in the elderly. These findings are similar to those of a previous study reporting positive associations between higher BPA concentrations and abnormal concentrations of liver enzymes (ie, γ-GTP, ALP and LDH) using National Health and Nutrition Examination Survey (NHANES) data in the USA.20

    The geometric mean of BPA levels in the present study was 0.58 μg/L compared to 1.33, 2.66, and 1.16 μg/L in the NHANES, German Environmental Surveys (GerES), and Canadian Health Measures Survey (CHMS), respectively.2 ,5 ,6

    We also calculated the ORs of liver function abnormality for urinary BPA concentrations divided by median concentrations. The high urinary BPA concentration group had twice the risk of liver function abnormality (OR 2.66; 95% CI 1.15 to 5.90) than the low urinary BPA concentration group.

    The major sources of human exposures to BPA are thought to be food and beverage consumption, because epoxy resins and polycarbonate plastics are closely related to those activities and contain substantial amounts of BPA.1 ,23–25 A recent study reported that urinary BPA concentrations increased more than 1000% in subjects who consumed one can of soup per day for 5 days compared to subjects who ate fresh soup.26 These results indicate that BPA leaches out of source materials in normal condition of use, which can be accelerated if the materials are exposed to high temperatures or acidic environments.27 ,28 Since plastics and cans are widely used, over 90% of human urine samples from adults and children contain considerable amounts of BPA.2 ,5 ,6 According to the recent large-scale NHANES data, daily BPA intake in men and women is estimated to be 49.1–59.0 and 36.2–46.5 ng kg−1 day−1, respectively.29 Although data on daily BPA intake were not available in the present study, the mean level of 1.13 μg/g creatinine as an indicator of urinary BPA is lower than 1.36 μg/g creatinine reported in the USA.2 ,6 Therefore, we assume that the present study subjects are not ingesting more BPA than Americans.

    Orally administered BPA is rapidly absorbed from the gastrointestinal tract and becomes free BPA which is the biologically active form. BPA is metabolised by the first-pass metabolism in liver to BPA-monoglucuronide and excreted in the urine in less than 6 h.9 ,30 Despite its rapid metabolism and elimination, the ubiquity of BPA exposure means a certain amount of BPA is maintained in the body most of the time. In other words, there is no washout period for BPA, because it is regularly supplied to the body via food and beverages that have been in contact with source materials. However, BPA concentrations in the body can change significantly depending on what was eaten recently.26

    Although BPA is known to induce damage to the human liver and cultured rat hepatocytes, the underlying mechanism has not been clarified.19 ,20 ,31 A possible mechanism is the direct toxicity of BPA to the liver. The liver is known to metabolise and remove most absorbed chemicals from the body. Therefore, the liver can be directly damaged by various chemicals including BPA. BPA was recently termed an ‘environmental obesogen,’ meaning that BPA can cause the development of obesity.32 BPA exposure can induce adipose cell enlargement or increase adipose cell number in vivo and in vitro.33 ,34 The action of BPA either directly triggers and accelerates the differentiation of fibroblasts into adipose cells or stimulates glucose transporter to increase glucose uptake followed by adipose cell enlargement. Additionally, BPA exposure may increase serum concentrations of adipokines (eg, interleukin-6 and tumour necrosis factor-α) and leptins, and inhibit adiponectin release, thereby inducing obesity.35–37 Therefore, adiposity caused by BPA exposure may contribute to fatty liver with elevated liver enzyme concentrations. In the present study, when we performed repeated data analysis using quartile of BPA and obesity (BMI <25 and ≥25), we observed a significant relationship although the relationship was not significant for the first visit data only in table 1 (see online supplementary table S6, supplementary figure S2).

    The strengths of the present study merit discussion. The panel study design with repeated measurements of serum and urine biomarkers for each subject provides a good opportunity for evaluating the effects of changes in BPA concentrations on liver function over time. We used mixed-effects models to assess the effects of visit-to-visit variations in BPA concentrations on liver function to account for intersubject and intrasubject variability. Individual exposure to other confounders would have biased the results only if they changed concomitantly with BPA concentrations, which is rather unlikely.

    However, this study has certain limitations. First, since our study subjects were elderly people, we cannot generalise the adverse health effects of BPA on the liver to all age groups. Second, although various possible confounding factors including age, sex, urinary cotinine, physical exercise, alcohol consumption, LDL-C level and BMI were adjusted for in the analytical models, other factors may have been overlooked. For instance, we did not measure biomarkers for another xenoestrogens, which may cause obesity and hepatocyte lipid accumulation. Third, although we excluded subjects who already had liver disease (eg, viral hepatitis, fatty liver disease and liver cancer), it is possible that subjects with mild liver dysfunction exhibited higher urinary BPA levels. Hepatic glucuronidation is the major metabolic pathway of the compound, and BPA is excreted into bile after glucuronidation. If a subject had liver dysfunction, they could excrete more BPA in urine than bile.38–40 Therefore, reverse relationships cannot be ruled out completely.

    In conclusion, urinary concentrations of BPA are associated with elevated liver enzyme levels in the elderly. If this association is causal, this study may have important implications regarding the control of BPA exposure in daily life, particularly in susceptible populations.

    What is already known on this subject

    • Recent evidence from animal studies shows that liver function is adversely affected by BPA treatment. A recent cross-sectional human study also shows an association between higher BPA concentrations and clinically abnormal concentrations of liver enzymes. However, the relationship between BPA exposure and liver function abnormality in humans has not been clarified.

    What this study adds

    • We found that community-level exposure to BPA is associated with abnormal liver function in the elderly, indicating that more stringent control of BPA is necessary to protect susceptible populations.

    • This result may be support policies which restrict BPA production and use especially in susceptible population.

    Acknowledgments

    We thank Yu-Mi Choi and Kyung-Hee Cho for assisting with data collection.

    References

    1. ↵
      1. Vandenberg LN,
      2. Hauser R,
      3. Marcus M,
      4. et al
      . Human exposure to bisphenol A (BPA). Reprod Toxicol 2007;24:139–77.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Bushnik T,
      2. Haines D,
      3. Levallois P,
      4. et al
      . Lead and bisphenol A concentrations in the Canadian population. Health Rep 2010;21:7–18.
      OpenUrlPubMed
    3. ↵
      1. Hanioka N,
      2. Takeda Y,
      3. Tanaka-Kagawa T,
      4. et al
      . Interaction of bisphenol A with human UDP-glucuronosyltransferase 1A6 enzyme. Environ Toxicol 2008;23:407–12.
      OpenUrlCrossRefPubMed
    4. ↵
      1. vom Saal FS,
      2. Hughes C
      . An extensive new literature concerning low-dose effects of bisphenol A shows the need for a new risk assessment. Environ Health Perspect 2005;113:926–33.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Becker K,
      2. Goen T,
      3. Seiwert M,
      4. et al
      . GerES IV: phthalate metabolites and bisphenol A in urine of German children. Int J Hyg Environ Health 2009;212:685–92.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Calafat AM,
      2. Kuklenyik Z,
      3. Reidy JA,
      4. et al
      . Urinary concentrations of bisphenol A and 4-nonylphenol in a human reference population. Environ Health Perspect 2005;113:391–5.
      OpenUrlPubMedWeb of Science
    7. ↵
      1. Ikezuki Y,
      2. Tsutsumi O,
      3. Takai Y,
      4. et al
      . Determination of bisphenol A concentrations in human biological fluids reveals significant early prenatal exposure. Human Reprod 2002;17:2839–41.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Schonfelder G,
      2. Wittfoht W,
      3. Hopp H,
      4. et al
      . Parent bisphenol A accumulation in the human maternal-fetal-placental unit. Environ Health Perspect 2002;110:A703–7.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Volkel W,
      2. Colnot T,
      3. Csanady GA,
      4. et al
      . Metabolism and kinetics of bisphenol a in humans at low doses following oral administration. Chem Res Toxicol 2002;15:1281–7.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Gould JC,
      2. Leonard LS,
      3. Maness SC,
      4. et al
      . Bisphenol A interacts with the estrogen receptor alpha in a distinct manner from estradiol. Mol Cell Endocrinol 1998;142:203–14.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Kuiper GG,
      2. Lemmen JG,
      3. Carlsson B,
      4. et al
      . Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor beta. Endocrinology 1998;139:4252–63.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Geens T,
      2. Aerts D,
      3. Berthot C,
      4. et al
      . A review of dietary and non-dietary exposure to bisphenol-A. Food Chem Toxicol 2012;50:3725–40.
      OpenUrlCrossRefPubMed
    13. ↵
      1. Howdeshell KL,
      2. Hotchkiss AK,
      3. Thayer KA,
      4. et al
      . Exposure to bisphenol A advances puberty. Nature 1999;401:763–4.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Newbold RR,
      2. Jefferson WN,
      3. Padilla-Banks E
      . Long-term adverse effects of neonatal exposure to bisphenol A on the murine female reproductive tract. Reprod Toxicol 2007;24:253–8.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Timms BG,
      2. Howdeshell KL,
      3. Barton L,
      4. et al
      . Estrogenic chemicals in plastic and oral contraceptives disrupt development of the fetal mouse prostate and urethra. Proc Natl Acad Sci USA 2005;102:7014–19.
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Ishido M,
      2. Yonemoto J,
      3. Morita M
      . Mesencephalic neurodegeneration in the orally administered bisphenol A-caused hyperactive rats. Toxicol Lett 2007;173:66–72.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Moriyama K,
      2. Tagami T,
      3. Akamizu T,
      4. et al
      . Thyroid hormone action is disrupted by bisphenol A as an antagonist. J Clin Endocrinol Metab 2002;87:5185–90.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Ropero AB,
      2. Alonso-Magdalena P,
      3. Garcia-Garcia E,
      4. et al
      . Bisphenol-A disruption of the endocrine pancreas and blood glucose homeostasis. Int J Androl 2008;31:194–200.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Korkmaz A,
      2. Ahbab MA,
      3. Kolankaya D,
      4. et al
      . Influence of vitamin C on bisphenol A, nonylphenol and octylphenol induced oxidative damages in liver of male rats. Food Chem Toxicol 2010;48:2865–71.
      OpenUrlCrossRefPubMed
    20. ↵
      1. Lang IA,
      2. Galloway TS,
      3. Scarlett A,
      4. et al
      . Association of urinary bisphenol A concentration with medical disorders and laboratory abnormalities in adults. JAMA 2008;300:1303–10.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Asahi J,
      2. Kamo H,
      3. Baba R,
      4. et al
      . Bisphenol A induces endoplasmic reticulum stress-associated apoptosis in mouse non-parenchymal hepatocytes. Life Sci 2010;87:431–8.
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Li YJ,
      2. Song TB,
      3. Cai YY,
      4. et al
      . Bisphenol A exposure induces apoptosis and upregulation of Fas/FasL and caspase-3 expression in the testes of mice. Toxicol Sci 2009;108:427–36.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Kang JH,
      2. Kondo F,
      3. Katayama Y
      . Human exposure to bisphenol A. Toxicology 2006;226:79–89.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Schecter A,
      2. Malik N,
      3. Haffner D,
      4. et al
      . Bisphenol A (BPA) in U.S. food. Environ Sci Technol 2010;44:9425–30.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Stahlhut RW,
      2. Welshons WV,
      3. Swan SH
      . Bisphenol A data in NHANES suggest longer than expected half-life, substantial nonfood exposure, or both. Environ Health Perspect 2009;117:784–9.
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Carwile JL,
      2. Ye X,
      3. Zhou X,
      4. et al
      . Canned soup consumption and urinary bisphenol A: a randomized crossover trial. JAMA 2011;306:2218–20.
      OpenUrlCrossRefPubMed
    27. ↵
      1. Brede C,
      2. Fjeldal P,
      3. Skjevrak I,
      4. et al
      . Increased migration levels of bisphenol A from polycarbonate baby bottles after dishwashing, boiling and brushing. Food Addit Contam 2003;20:684–9.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Carwile JL,
      2. Luu HT,
      3. Bassett LS,
      4. et al
      . Polycarbonate bottle use and urinary bisphenol A concentrations. Environ Health Perspect 2009;117:1368–72.
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Lakind JS,
      2. Naiman DQ
      . Bisphenol A (BPA) daily intakes in the United States: estimates from the 2003–2004 NHANES urinary BPA data. J Expo Sci Environ Epidemiol 2008;18:608–15.
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Domoradzki JY,
      2. Pottenger LH,
      3. Thornton CM,
      4. et al
      . Metabolism and pharmacokinetics of bisphenol A (BPA) and the embryo-fetal distribution of BPA and BPA-monoglucuronide in CD Sprague-Dawley rats at three gestational stages. Toxicol Sci 2003;76:21–34.
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Nakagawa Y,
      2. Tayama S
      . Metabolism and cytotoxicity of bisphenol A and other bisphenols in isolated rat hepatocytes. Arch Toxicol 2000;74:99–105.
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Grun F
      . Obesogens. Curr Opin Endocrinol Diabetes Obes 2010;17:453–9.
      OpenUrlCrossRefPubMed
    33. ↵
      1. Masuno H,
      2. Iwanami J,
      3. Kidani T,
      4. et al
      . Bisphenol a accelerates terminal differentiation of 3T3-L1 cells into adipocytes through the phosphatidylinositol 3-kinase pathway. Toxicol Sci 2005;84:319–27.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Masuno H,
      2. Kidani T,
      3. Sekiya K,
      4. et al
      . Bisphenol A in combination with insulin can accelerate the conversion of 3T3-L1 fibroblasts to adipocytes. J Lipid Res 2002;43:676–84.
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Ben-Jonathan N,
      2. Hugo ER,
      3. Brandebourg TD
      . Effects of bisphenol A on adipokine release from human adipose tissue: Implications for the metabolic syndrome. Mol Cell Endocrinol 2009;304:49–54.
      OpenUrlCrossRefPubMed
    36. ↵
      1. Hugo ER,
      2. Brandebourg TD,
      3. Woo JG,
      4. et al
      . Bisphenol A at environmentally relevant doses inhibits adiponectin release from human adipose tissue explants and adipocytes. Environ Health Perspect 2008;116:1642–7.
      OpenUrlCrossRefPubMedWeb of Science
    37. ↵
      1. Miyawaki J,
      2. Sakayama K,
      3. Kato H,
      4. et al
      . Perinatal and postnatal exposure to bisphenol a increases adipose tissue mass and serum cholesterol level in mice. J Atheroscler Thromb 2007;14:245–52.
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      1. Pritchett JJ,
      2. Kuester RK,
      3. Sipes IG
      . Metabolism of bisphenol a in primary cultured hepatocytes from mice, rats, and humans. Drug Metab Dispos 2002;30:1180–5.
      OpenUrlAbstract/FREE Full Text
    39. ↵
      1. Matthews JB,
      2. Twomey K,
      3. Zacharewski TR
      . In vitro and in vivo interactions of bisphenol A and its metabolite, bisphenol A glucuronide, with estrogen receptors alpha and beta. Chem Res Toxicol 2001;14:149–57.
      OpenUrlCrossRefPubMedWeb of Science
    40. ↵
      1. Inoue H,
      2. Yokota H,
      3. Makino T,
      4. et al
      . Bisphenol a glucuronide, a major metabolite in rat bile after liver perfusion. Drug Metab Dispos 2001;29:1084–7.
      OpenUrlAbstract/FREE Full Text

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • MRL and HP contributed equally.

    • Contributors HP, MRL and YCH planned the study, analysed data and wrote the manuscript. MR Lee and YH Lim contributed to statistical analyses. SHB, JHK and SHC were responsible for interviewing participants and cleaning the collected data.

    • Funding This study was supported by the Environmental Exposures and Health Effects in Elderly Population programme (0411-20080013, 0411-20090007, 0411-20100016) of the Ministry of Environment, Republic of Korea.

    • Competing interests None.

    • Patient consent Obtained.

    • Ethics approval The Institutional Review Board of the Seoul National University Hospital.

    • Provenance and peer review Not commissioned; externally peer reviewed.