Abstract
Background:
Bisphenol A (BPA) is a widely used chemical in the manufacture of polycarbonate plastics and epoxy resins. Recent animal studies have suggested that BPA exposure may have a role in the development of weight gain, insulin resistance, pancreatic endocrine dysfunction, thyroid hormone disruption, and several other mechanisms involved in the development of diabetes. However, few human studies have examined the association between markers of BPA exposure and diabetes mellitus.
Methods:
We examined the association between urinary BPA levels and diabetes mellitus in the National Health and Nutritional Examination Survey (NHANES) 2003–2008. Urinary BPA levels were examined in quartiles. The main outcome of interest was diabetes mellitus defined according the latest American Diabetes Association guidelines.
Results:
Overall, we observed a positive association between increasing levels of urinary BPA and diabetes mellitus, independent of confounding factors such as age, gender, race/ethnicity, body mass index, and serum cholesterol levels. Compared to quartile 1 (referent), the multivariate-adjusted odds ratio (95% confidence interval) of diabetes associated with quartile 4 was 1.68 (1.22–2.30) (p-trend = 0.002). The association was present among normal-weight as well as overweight and obese subjects.
Conclusions:
Urinary BPA levels are found to be associated with diabetes mellitus independent of traditional diabetes risk factors. Future prospective studies are needed to confirm or disprove this finding.
Bisphenol A (BPA) is a chemical produced in very high volume, with more than 2 million metric tons produced worldwide in 2003 (1). It is used extensively in the manufacture of epoxy resins, polycarbonate plastics, and food and beverage containers (1), and detectable levels of BPA in urine have been shown to be present in the majority of U.S. adults (2, 3). BPA has been shown to have estrogenic and thyroid hormone-disrupting effects (1, 4, 5). Recent evidence, especially from animal studies, suggests that BPA exposure may have a role in weight gain, development of obesity and insulin resistance, and the subsequent development of diabetes mellitus (6, 7). However, the association between BPA levels and diabetes mellitus in humans is not clear.
Two previous studies (8, 9) have examined the association between urinary BPA levels and diabetes and reported opposing findings—with one study reporting a positive association (8) and the other not detecting an association (9). However, both of the studies (8, 9) used self-reported diabetes as the outcome; fasting glucose or glycosylated hemoglobin was not used as a criterion in defining diabetes in these studies as recommended by recent guidelines (10). Studies (11) have shown that, when using self-reported diabetes as opposed to blood glucose measurements, there is likely to be substantial misclassification of diabetes status and this may consequently lead to biased risk associations. In this context, we examined the association between urinary BPA levels and diabetes mellitus in the 2003 to 2008 National Health and Nutritional Examination Survey (NHANES), a representative sample of U.S. adults. In the current report we had fasting glucose levels as well as glycosylated hemoglobin to define diabetes mellitus according to the latest American Diabetes Association guidelines.
Subjects and Methods
The current study is based on data from the NHANES 2003–2008. Detailed descriptions of NHANES study design and methods are available elsewhere (12–14). In brief, the NHANES survey included a stratified multistage probability sample representative of the civilian noninstitutionalized U.S. population. Selection was based on counties, blocks, households, and individuals within households and included the oversampling of non-Hispanic blacks and Mexican-Americans to provide stable estimates of these groups. Subjects were required to sign a consent form before their participation, and approval was obtained from the Human Subjects Committee of the U.S. Department of Health and Human Services.
The current study sample consisted of participants more than 20 yr old among whom urinary BPA was available (n = 4792). We excluded subjects with self-reported cardiovascular disease (n = 495) and also subjects with missing data (n = 330) on covariates included in the multivariable model, including level of education, smoking status, serum or fasting glucose levels, systolic or diastolic blood pressure, body mass index (BMI), or cholesterol levels. This resulted in 3967 participants (51.7% women), 467 of whom had diabetes.
Exposure measurements
Age, gender, race/ethnicity, smoking status, alcohol intake (grams per day), level of education, history of diabetes, and oral hypoglycemic intake or insulin administration were assessed using a questionnaire (12–14). Individuals who had not smoked more than 100 cigarettes in their lifetime were considered never smokers; those who had smoked more than 100 cigarettes in their lifetime were considered former smokers if they answered negatively to the question “Do you smoke now?” and current smokers if they answered affirmatively. BMI was calculated as weight in kilograms divided by height in meters squared.
Rigorous procedures with quality control checks were used in blood collection, and details about these procedures are provided in the NHANES Laboratory/Medical Technologists Procedures Manual (12–14). Seated systolic and diastolic blood pressures were measured using a mercury sphygmomanometer according to the American Heart Association and Seventh Joint National Committee recommendations (15). Up to three measurements were averaged for systolic and diastolic blood pressures. Patients were considered hypertensive if they reported the current use of blood pressure-reducing medication and/or had systolic blood pressure greater than 140 mm Hg and/or diastolic blood pressure greater than 90 mm Hg (15). Urinary creatinine was analyzed using the Jaffe rate reaction method and using the CX3 analyzer (Beckman Coulter, Inc., Brea, CA) (16).
Previous measures of BPA in biological matrixes involved techniques such as gas chromatography (GC) or HPLC (17). To achieve enhanced sensitivity and selectivity over previous methods, in the current NHANES, measures of environmental phenols were derivatized to alkyl or acyl derivatives before GC/mass spectrometry analysis (16). Using solid-phase extraction coupled to HPLC and tandem mass spectrometry, detection levels of 0.1–2 ng/ml in 100 μl of urine were achieved, sufficient for measuring urinary BPA levels in non-occupationally exposed participants (16).
Main outcome of interest: diabetes
Serum glucose was measured using the modified hexokinase method at the University of Missouri Diabetes Diagnostic Laboratory. Diabetes mellitus was defined based on the recent guidelines of the American Diabetes Association (10) as a serum glucose greater than 126 mg/dl after fasting for a minimum of 8 h, a serum glucose greater than 200 mg/dl for those who fasted less than 8 h before their NHANES visit, a glycosylated hemoglobin value greater than 6.5%, or self-reported current use of oral hypoglycemic medication or insulin.
Statistical analysis
Urinary BPA was categorized into quartiles (<1.10, 1.10–2.10, 2.11–4.20, and >4.20 ng/ml). We hypothesized that high BPA levels are associated with diabetes mellitus. The odds ratio (OR) [95% confidence interval (CI)] of diabetes for BPA was calculated by taking the lowest quartile (quartile 1) as the referent and using multivariable logistic regression models. We used two models: the age- and sex-adjusted model and the multivariable model, additionally adjusting for race/ethnicity (non-Hispanic whites, non-Hispanic blacks, Mexican-Americans, and others), education categories (below high school, high school, above high school), smoking (never smoker, former smoker, current smoker), alcohol intake (nondrinker, moderate drinker, heavy drinker), BMI (normal, overweight, obese), systolic blood pressure, diastolic blood pressure, urinary creatinine (milligrams per deciliter), and total serum cholesterol (milligrams per deciliter). Trends in the OR of diabetes across increasing urinary BPA categories were determined by modeling BPA as an ordinal variable. Sample weights that account for the unequal probabilities of selection, oversampling, and nonresponse were applied for all analyses using SAS (version 9.2; SAS Institute, Cary, NC) and SUDAAN software; se values were estimated using the Taylor series linearization method.
Results
Table 1 shows the baseline characteristics of the population by gender. Smokers were found to have a statistically nonsignificant higher mean BPA level (4.17 ± 0.29 ng/dl) when compared with nonsmokers (3.86 ± 0.19 ng/dl) (P = 0.216). Table 2 shows the association between increasing levels of BPA and diabetes mellitus. Overall, we observed positive association between increasing BPA levels and diabetes in both the age- and sex-adjusted model and the multivariable-adjusted model. Models evaluating trend in this association were also statistically significant.
Table 1.
Baseline characteristics of the study population by gender
| Characteristics | Men | Women |
|---|---|---|
| n | 1879 | 2088 |
| Age (yr) | 44.3 ± 0.5 | 45.6 ± 0.4 |
| Race/ethnicity | ||
| Non-Hispanic whites | 953 (71.5) | 991 (70.0) |
| Non-Hispanic blacks | 373 (9.7) | 425 (11.5) |
| Mexican-Americans | 351 (8.5) | 436 (8.0) |
| Others | 202 (10.2) | 236 (10.4) |
| Education categories | ||
| Below high school | 511 (17.9) | 560 (16.7) |
| High school | 476 (25.5) | 510 (24.9) |
| Above high school | 892 (56.6) | 1018 (58.4) |
| Smoking | ||
| Never smoker | 840 (45.7) | 1289 (58.7) |
| Former smoker | 529 (27.2) | 403 (20.3) |
| Current smoker | 510 (27.1) | 396 (21.0) |
| Alcohol intake | ||
| Nondrinker | 533 (24.5) | 894 (35.7) |
| Moderate drinker | 663 (38.7) | 863 (47.3) |
| Heavy drinker | 683 (36.8) | 331 (17.0) |
| BMI (kg/m2) | ||
| Normal weight (<25) | 544 (28.6) | 695 (40.3) |
| Overweight (25–30) | 726 (38.1) | 618 (26.5) |
| Obese (BMI ≥ 30) | 609 (33.3) | 775 (33.2) |
| Systolic blood pressure (mm Hg) | 123.1 ± 0.4 | 119.1 ± 0.5 |
| Diastolic blood pressure (mm Hg) | 72.4 ± 0.4 | 69.2 ± 0.3 |
| Urinary creatinine (mg/dl) | 147.38 ± 2.63 | 106.55 ± 2.14 |
| Total cholesterol (mg/dl) | 201.55 ± 1.12 | 202.23 ± 1.03 |
| Urinary BPA (ng/ml) | 3.97 ± 0.21 | 3.90 ± 0.26 |
| Diabetes (%) | 229 (9.5) | 238 (8.7) |
Data are presented as weighted number (percentage) or mean ± se by gender, as appropriate for the variable.
Table 2.
Association between urinary BPA and diabetes mellitus
| BPA quartiles (ng/ml) | Sample size (diabetes %) | Age-, sex-adjusted, OR (95% CI) | Multivariable-adjusted, OR (95% CI)a |
|---|---|---|---|
| Quartile 1 (<1.10) | 1121 (8.3) | 1 (referent) | 1 (referent) |
| Quartile 2 (1.10–2.10) | 905 (10.8) | 1.55 (1.19–2.02) | 1.42 (1.03–1.96) |
| Quartile 3 (2.11–4.20) | 977 (11.2) | 1.60 (1.25–2.05) | 1.48 (1.05–2.08) |
| Quartile 4 (>4.20) | 964 (12.8) | 1.81 (1.36–2.43) | 1.68 (1.22–2.30) |
| p-trend | <0.0001 | 0.002 |
Adjusted for age (years), gender, race-ethnicity (non-Hispanic whites, non-Hispanic blacks, Mexican-Americans, others), education categories (below high school, high school, above high school), smoking (never, former, current), alcohol intake (never, former, current), BMI (normal, overweight, obese), systolic and diastolic blood pressure (mm Hg), urinary creatinine (mg/dl), and total cholesterol (mg/dl).
Table 3 shows the association between increasing BPA levels and diabetes mellitus by BMI categories. We found that the association between increasing BPA levels and diabetes was consistently present among normal-weight as well as overweight/obese subjects; p-trends for the association were also significant. In a supplementary analysis, we examined the association between BPA and diabetes separately among smokers and nonsmokers. We found that, consistent with our main findings in Table 2, there was a positive association between BPA and diabetes among smokers and nonsmokers; the p-interaction for the cross-product BPA × smoking status term was 0.6530, suggesting that the association between BPA and diabetes did not differ by smoking status.
Table 3.
Association between urinary BPA and diabetes mellitus by BMI
| BPA quartiles (ng/ml) | Normal weight |
Overweight/obese |
||
|---|---|---|---|---|
| Sample size | Multivariable-adjusted, OR (95% CI)a | Sample size | Multivariable-adjusted, OR (95% CI)a | |
| Quartile 1 (<1.10) | 408 | 1 (referent) | 713 | 1 (referent) |
| Quartile 2 (1.10–2.10) | 276 | 2.75 (1.03–7.33) | 629 | 1.27 (0.90–1.79) |
| Quartile 3 (2.11–4.20) | 272 | 2.14 (0.79–5.81) | 705 | 1.41 (1.00–1.98) |
| Quartile 4 (>4.20) | 283 | 3.17 (1.23–8.18) | 681 | 1.56 (1.09–2.24) |
| p-trend | 0.03 | 0.01 | ||
Adjusted for age (years), gender, race-ethnicity (non-Hispanic whites, non-Hispanic blacks, Mexican-Americans, others), education categories (below high school, high school, above high school), smoking (never, former, current), alcohol intake (never, former, current), systolic and diastolic blood pressure (mm Hg), urinary creatinine (mg/dl), and total cholesterol (mg/dl).
Discussion
In a large multiethnic, nationally representative sample, we found that increasing serum BPA levels are positively associated with diabetes mellitus. The observed association was found to be independent of confounding factors such as BMI, urinary creatinine (18), alcohol intake, and serum cholesterol level. Our study adds to the emerging evidence of the role of environmental exposure to BPA on cardiometabolic health in humans.
BPA is an environmental chemical used as a constituent monomer in polycarbonate plastics, which are used extensively in drink containers and food packaging and in the production of oxidant used in the lining of canned goods (1). Exposure to BPA is believed to be mainly through dietary intake, with additional exposure through water, dental sealants, inhalation of household dusts, and exposure through skin (1). Recent studies have documented that over 90% of the U.S. general population has measurable concentrations of BPA metabolites in urine (2, 3).
Several lines of recent evidence suggest that an association between urinary BPA levels and diabetes mellitus may be biologically plausible. Animal studies have shown that BPA exposure may have a role in weight gain and obesity development through several mechanisms, including the action of BPA on preadipocytes (19, 20), a role as an estrogen (6), potential interactions with estrogen-related receptor γ (21), action as a thyroid hormone antagonist (4), role as a peroxisome proliferator-activated receptor γ antagonist (22), and its role in influencing pancreatic endocrine function (23). Furthermore, Alonso-Magdalena et al. (24) in a recent experiment showed that mice exposed to BPA levels as low as 10 μg/kg·d developed hyperinsulinemia, insulin resistance, and glucose intolerance. Finally, Carwile and Michels (25) showed that urinary BPA levels are associated with obesity in the NHANES survey. Therefore, it is possible that BPA may contribute to obesity and thereby indirectly to the development of type 2 diabetes.
However, there are few studies in humans for comparison. Two previous studies (8, 9) have examined the association between higher BPA levels and self-reported diabetes and have reported conflicting results; whereas one study found a positive association (8), the other did not (9). However, it is well known that in epidemiological studies (11), self-reported diabetes largely underdiagnoses the actual prevalence of diabetes in a population. When using self-reported diabetes, a substantial number of people who actually have diabetes may be misclassified as normoglycemic, and such a misclassification would likely underestimate a true association if it is present. The main advantage of our study over the previous two studies is that we defined diabetes consistent with the latest American Diabetes Association guidelines (10) and included fasting glucose, nonfasting glucose, and glycosylated hemoglobin levels in addition to self-reported diabetes. Consequently, we found that there was a positive association between serum BPA levels and diabetes mellitus in this nationally representative sample of U.S. adults. In subsequent stratified analysis, the observed association was found to be present among both normal-weight and overweight/obese adults.
The main strengths of our study include its nationally representative sample, use of rigorous study methods to collect the data, and the availability of extensive data on confounders. The main study limitation is that the current study is cross-sectional in nature, therefore making it impossible to draw cause and effect in the observed associations. Future prospective studies are required to confirm or disprove our findings.
In summary, we found that in a nationally representative sample of U.S. adults, higher BPA levels were positively associated with diabetes mellitus independent of confounding factors such as age, BMI, alcohol intake, and cholesterol levels. If confirmed in future prospective studies, reducing environmental exposure to BPA may have a role in the prevention of diabetes mellitus.
Acknowledgments
This study was funded by an American Heart Association National Clinical Research Program grant (to A.S.) and National Institutes of Health Grant 5R03ES018888-02 (to A.S.).
Both authors contributed to the intellectual development of this paper. A.S. had the original idea for the study, wrote the paper, and is the guarantor. S.T. performed the statistical analyses and was involved in critical revisions to the manuscript.
Disclosure Summary: There are no conflicts of interest related to this manuscript.
Footnotes
- BMI
- Body mass index
- BPA
- bisphenol A
- CI
- confidence interval
- GC
- gas chromatography
- OR
- odds ratio.
References
- 1. Vandenberg LN, Maffini MV, Sonnenschein C, Rubin BS, Soto AM. 2009. Bisphenol-A and the great divide: a review of controversies in the field of endocrine disruption. Endocr Rev 30:75–95 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Calafat AM, Kuklenyik Z, Reidy JA, Caudill SP, Ekong J, Needham LL. 2005. Urinary concentrations of bisphenol A and 4-nonylphenol in a human reference population. Environ Health Perspect 113:391–395 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Calafat AM, Ye X, Wong LY, Reidy JA, Needham LL. 2008. Exposure of the U.S. population to bisphenol A and 4-tertiary-octylphenol: 2003–2004. Environ Health Perspect 116:39–44 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Moriyama K, Tagami T, Akamizu T, Usui T, Saijo M, Kanamoto N, Hataya Y, Shimatsu A, Kuzuya H, Nakao K. 2002. Thyroid hormone action is disrupted by bisphenol A as an antagonist. J Clin Endocrinol Metab 87:5185–5190 [DOI] [PubMed] [Google Scholar]
- 5. Zoeller RT. 2005. Environmental chemicals as thyroid hormone analogues: new studies indicate that thyroid hormone receptors are targets of industrial chemicals? Mol Cell Endocrinol 242:10–15 [DOI] [PubMed] [Google Scholar]
- 6. Rubin BS, Soto AM. 2009. Bisphenol A: perinatal exposure and body weight. Mol Cell Endocrinol 304:55–62 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Newbold RR, Padilla-Banks E, Jefferson WN. 2009. Environmental estrogens and obesity. Mol Cell Endocrinol 304:84–89 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Lang IA, Galloway TS, Scarlett A, Henley WE, Depledge M, Wallace RB, Melzer D. 2008. Association of urinary bisphenol A concentration with medical disorders and laboratory abnormalities in adults. JAMA 300:1303–1310 [DOI] [PubMed] [Google Scholar]
- 9. Melzer D, Rice NE, Lewis C, Henley WE, Galloway TS. 2010. Association of urinary bisphenol a concentration with heart disease: evidence from NHANES 2003/06. PLoS One 5:e8673. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. American Diabetes Association 2011. Diagnosis and classification of diabetes mellitus. Diabetes Care 34(Suppl 1):S62–S69 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Molenaar EA, Van Ameijden EJ, Grobbee DE, Numans ME. 2007. Comparison of routine care self-reported and biometrical data on hypertension and diabetes: results of the Utrecht Health Project. Eur J Public Health 17:199–205 [DOI] [PubMed] [Google Scholar]
- 12.National Center for Health Statistics. National Health and Nutrition Examination Survey: survey operations manuals. 2003–2004. [Accessed September 14, 2011]. [article available at: http://www.cdc.gov/nchs/nhanes/nhanes2003-2004/current_nhanes_03_04.htm]
- 13.National Center for Health Statistics. National Health and Nutrition Examination Survey: survey operations manuals. 2005–2006. [Accessed September 14, 2011]. [article available at: http://www.cdc.gov/nchs/nhanes/nhanes2005-2006/current_nhanes_05_06.htm]
- 14.National Center for Health Statistics. National Health and Nutrition Examination Survey: survey operations manuals. 2007–2008. [Accessed September 14, 2011]. [article available at: http://www.cdc.gov/nchs/nhanes/nhanes2007-2008/current_nhanes_07_08.htm]
- 15. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL, Jr, Jones DW, Materson BJ, Oparil S, Wright JT, Jr, Roccella EJ, Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure; National Heart, Lung, and Blood Institute; National High Blood Pressure Education Program Coordinating Committee 2003. Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension 42:1206–1252 [DOI] [PubMed] [Google Scholar]
- 16.National Center for Health Statistics. National Health and Nutrition Examination Survey: laboratory procedures. 2003–2004. [Accessed September 14, 2011]. [article available at: http://www.cdc.gov/nchs/data/nhanes/nhanes_03_04/lab_c_generaldoc.pdf]
- 17. Ye X, Kuklenyik Z, Needham LL, Calafat AM. 2005. Automated on-line column-switching HPLC-MS/MS method with peak focusing for the determination of nine environmental phenols in urine. Anal Chem 77:5407–5413 [DOI] [PubMed] [Google Scholar]
- 18. Barr DB, Wilder LC, Caudill SP, Gonzalez AJ, Needham LL, Pirkle JL. 2005. Urinary creatinine concentrations in the U.S. population: implications for urinary biologic monitoring measurements. Environ Health Perspect 113:192–200 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Phrakonkham P, Viengchareun S, Belloir C, Lombès M, Artur Y, Canivenc-Lavier MC. 2008. Dietary xenoestrogens differentially impair 3T3-L1 preadipocyte differentiation and persistently affect leptin synthesis. J Steroid Biochem Mol Biol 110:95–103 [DOI] [PubMed] [Google Scholar]
- 20. Masuno H, Iwanami J, Kidani T, Sakayama K, Honda K. 2005. Bisphenol A accelerates terminal differentiation of 3T3-L1 cells into adipocytes through the phosphatidylinositol 3-kinase pathway. Toxicol Sci 84:319–327 [DOI] [PubMed] [Google Scholar]
- 21. Matsushima A, Teramoto T, Okada H, Liu X, Tokunaga T, Kakuta Y, Shimohigashi Y. 2008. ERRγ tethers strongly bisphenol A and 4-α-cumylphenol in an induced-fit manner. Biochem Biophys Res Commun 373:408–413 [DOI] [PubMed] [Google Scholar]
- 22. Wright HM, Clish CB, Mikami T, Hauser S, Yanagi K, Hiramatsu R, Serhan CN, Spiegelman BM. 2000. A synthetic antagonist for the peroxisome proliferator-activated receptor γ inhibits adipocyte differentiation. J Biol Chem 275:1873–1877 [DOI] [PubMed] [Google Scholar]
- 23. Ropero AB, Alonso-Magdalena P, García-García E, Ripoll C, Fuentes E, Nadal A. 2008. Bisphenol-A disruption of the endocrine pancreas and blood glucose homeostasis. Int J Androl 31:194–200 [DOI] [PubMed] [Google Scholar]
- 24. Alonso-Magdalena P, Morimoto S, Ripoll C, Fuentes E, Nadal A. 2006. The estrogenic effect of bisphenol A disrupts pancreatic β-cell function in vivo and induces insulin resistance. Environ Health Perspect 114:106–112 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Carwile JL, Michels KB. 2011. Urinary bisphenol A and obesity: NHANES 2003–2006. Environ Res 111:825–830 [DOI] [PMC free article] [PubMed] [Google Scholar]