Effects of bisphenol A on cardiovascular disease: An epidemiological study using National Health and Nutrition Examination Survey 2003–2016 and meta-analysis

• Published in: The Science of the total environment Vol. 763; p. 142941
• Main Authors: Moon, Shinje, Yu, Sung Hoon, Lee, Chang Beom, Park, Young Joo, Yoo, Hyung Joon, Kim, Dong Sun
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.04.2021
• Subjects: Adult; Benzhydryl Compounds; Bisphenol A; body mass index; Cardiovascular disease; Cardiovascular Diseases - chemically induced; Cardiovascular Diseases - epidemiology; cigarettes; confidence interval; creatinine; diabetes mellitus; dose response; Endocrine disrupting chemical; endocrine-disrupting chemicals; environment; Epidemiologic Studies; epidemiological studies; Humans; hyperlipidemia; hypertension; meta-analysis; National Health and Nutrition Examination Survey; nationalities and ethnic groups; Nutrition Surveys; obesity; odds ratio; Phenols; regression analysis; urine; Bisphenol A; Cardiovascular disease; Endocrine disrupting chemical
• ISSN: 0048-9697; 1879-1026; 1879-1026
• DOI: 10.1016/j.scitotenv.2020.142941

As the most widely consumed endocrine-disrupting chemical, bisphenol A (BPA) has been linked to reproductive dysfunction, diabetes mellitus, and obesity. However, the evidence for an association between BPA and cardiovascular disease (CVD) remains insufficient. In the present study, we aimed to identify the association between BPA and CVD, using data from the 2003–2016 National Health and Nutrition Examination Surveys (NHANES). We estimated urine BPA concentration after adjustments for creatinine (ng/mg) and normalized the asymmetrical distribution using natural logarithmic transformation (ln-BPA/Cr). A multivariate logistic regression was performed to evaluate the odds ratio (OR) and 95% confidence interval (CI) for CVD, with ln-BPA/Cr concentration as predictor. We then performed a Mantel–Haenszel meta-analysis with five eligible studies and NHANES 2003–2016 data. Our subjects were 11,857 adults from the NHANES data. After adjusting for age, sex, race/ethnicity, body mass index (BMI), cigarette smoking, diabetes status, hypertension, and dyslipidemia, OR between ln-BPA/Cr and CVD was 1.13 (95% CI: 1.02–1.24). After propensity-score-matching with age, sex, race/ethnicity, BMI, cigarette smoking, diabetes, hypertension, and dyslipidemia, OR continued to be significant for the association between ln-BPA/Cr and CVD (OR: 1.18, 95% CI: 1.04–1.33). A restricted cubic spline plot of this relationship revealed a dose-dependent increase in OR. However, untransformed BPA had a linear relationship with CVD only at low concentrations, whereas the OR of BPA plateaued at high concentrations. In a meta-analysis with 22,878 subjects, after adjusting for age, sex, and various cardiometabolic risk factors, OR was 1.13 (95% CI, 1.03–1.23). In conclusion, our study provides additional epidemiological evidence supporting an association between BPA and CVD. [Display omitted] •Association of bisphenol A and cardiovascular disease (CVD) has limited evidence.•Thus, we performed logistic regression using 2003–2016 NHANES and a meta-analysis.•In the 2003–2016 NHANES, bisphenol A was associated with CVD.•The meta-analysis also confirmed a significant association of bisphenol A with CVD.