Bisphenol A (BPA) is an industrial chemical used in the manufacturing of polycarbonate plastics for various consumer goods as food containers and epoxy resins for adhesives, paints, primers, and sealers1. Calafat et al. U.S. population-based study showed 92.6% prevalence of BPA exposure2. Observational studies on the adverse health effects of BPA have classified BPA as an endocrine disruptor and suggested association between BPA exposure and cardiovascular diseases including cardiomyopathies and cardiac arrhythmias3,4. Analysis of cross-sectional data collected by National Health and Nutrition Examination Survey 2003-2004 showed association between higher urinary BPA concentrations and reported angina, coronary artery disease (CAD), heart attack, diabetes mellitus (DM), and abnormal liver function5. As per the American Heart Association guidelines, dilated cardiomyopathy (DCM) is a group of genetic, environmental, or unknown disorders affecting the myocardium characterized by progressive ventricular dilatation and systolic dysfunction in absence of hypertension, valvular, congenital, or ischemic heart disease6. DCM has an estimated prevalence of 40 patients per 100,000 population7. Clinically, patients with DCM present with congestive heart failure (HF). In 2015, Xiong et al showed a highly significant difference in the serum BPA level of a cohort of DCM patients vs a cohort of healthy controls (t(174) = 3.1, P = < 0.001)8. There are few published studies which address the prevalence of BPA exposure and temporality of association in patients with DCM.
The study evaluated the prevalence of BPA exposure in patients with DCM versus healthy controls. Secondary objectives included assessment of the difference in the echocardiographic parameters of patients with DCM exposed to BPA versus patients with DCM not exposed to BPA.
Our study was a 1-year cross-sectional, single tertiary care cardiac centre, open-labeled, randomized, controlled study. The study design was approved by the hospital ethics committee and all participants signed written informed consents.
Study participants were subjected to history taking and data collection for age, gender, risk factors including hypertension, DM, smoking and family history, comprehensive clinical examination,12 leads electrocardiogram (ECG) to detect ECG signs of left atrial (LA) abnormalities, CAD, and/or pulmonary hypertension (PH), two-dimensional, motion-mode and Doppler echocardiography using Siemens acuson-70® to measure LA, left ventricular (LV), and right ventricular (RV) size, assess LV and RV function, and screen for PH, and laboratory investigations including assessment of serum BPA level. Data documented with ECG include atrial fibrillation, left bundle branch block and right bundle branch block, while data documented with echocardiography include LA diameter, LV end diastolic diameter (LVEDD), LV end systolic diameter (LVESD), ejection fraction (EF), ratio of early diastolic mitral inflow velocity to early diastolic mitral annular velocity, tricuspid annular plane systolic excursion (TAPSE), and pulmonary artery systolic pressure. Screened participants were excluded if they had ischemic cardiomyopathy, valvular or congenital heart diseases, thyroid, pituitary or adrenal disorders, end stage renal or hepatic failure, or received hormonal treatment.
Eighty eligible participants from in-and-out patients were randomly, consecutively assigned with an unequal allocation ratio into an open-labeled unblinded fashion. The enrolled study participants were divided into 2 groups of 50 participants with DCM and 30 age and gender matched asymptomatic participants with normal echocardiographic findings. In our study, we defined DCM by the symptoms of HF and the echocardiographic findings of dilated LV diameter (LVEDD of > 5.5 and LVESD of > 4.5) and impaired LV systolic function (EF of < 50%).
The study evaluated the prevalence of BPA exposure in patients with DCM versus healthy controls. Secondary objectives included assessment of the difference in the echocardiographic parameters of patients with DCM exposed to BPA versus patients with DCM not exposed to BPA.
The echocardiographic assessment outcomes were coded, and the data was analysed with the Statistical Package for the Social Sciences software (SPSS®) version 25. Quantitative data was expressed as means and standard deviations, while qualitative data was expressed as medians and ranges. Comparisons between parametrically distributed quantitative variables were done with independent t-test, between non-parametrically distributed quantitative variables with Mann-Whitney test, and between qualitative variables with Chi-square test, respectively9,10. The confidence interval was set to 95% and the margin of error accepted was set to 5%. Any comparison considered statistically significant was at P < 0.05 or less and highly significant at P < 0.01. Final data analysis was as per Intention to treat analysis.
We recruited 80 patients from one centre in one country from 2015 through 2016. The two study groups were balanced with regards to baseline socio-demographic characteristics and risk factors (Table 1). The key sociodemographic feature of enrolled participants was male gender predominance (74% of DCM group and 76.7% of healthy control group were males). Age was not significantly different between both groups (mean age was 56.7 ± 12.8 years for DCM group vs 58.2 ± 9.8 years for healthy control group, P = 0.568). Risk factors as hypertension, DM, smoking, and family history (variables believed to cause confounding) were equally distributed (matched) among both studied groups to adjust for confounding11. All enrolled participants completed the study and there were no withdrawals.
Comparison between dilated cardiomyopathy and healthy controls groups regarding the baseline characteristics and risk factors.
| Dilated Cardiomyopathy Group No: 50 | Healthy Controls Group No: 30 | Independent t-test | P value | ||
|---|---|---|---|---|---|
| Age Mean ± SD | 56.7 ± 12.8 | 58.2 ± 9.8 | 1.5 | 0.568 | |
| Dilated Cardiomyopathy Group | Healthy Controls Group | Chi-square test | P value | ||
| No: 50 | No: 30 | ||||
| Sex | Females | 13 (26.0%) | 7 (23.3%) | 0.07 | 0.790 |
| Males | 37 (74.0%) | 23 (76.7%) | |||
| Residence | Fayoum | 17 (34.0%) | 10 (33.0%) | 0.004 | 0.951 |
| Other Districts | 33 (66.0%) | 20 (66.7%) | |||
| 0 | 2 (4.0%) | 0 (0.0%) | |||
| No: of Offspring | 2 | 12 (24.0%) | 5 (16.7%) | 1.99 | 0.574 |
| 3 | 13 (26.0%) | 9 (30.0%) | |||
| ≥4 | 23 (46.0%) | 16 (53.3%) | |||
| Diabetes Mellitus | Yes | 9 (18.0%) | 8 (26.7%) | 0.84 | 0.359 |
| No | 41 (82.0%) | 22 (73.3%) | |||
| Hypertension | Yes | 14 (28.0%) | 9 (30.0%) | 0.0366 | 0.848 |
| No | 36 (72.0%) | 21 (70.0%) | |||
| Ex-smokers | 15 (30.0%) | 8 (26.7%) | |||
| Smoking | Smokers | 3 (6.0%) | 5 (16.7%) | 2.37 | 0.306 |
| Non-smokers | 32 (64.0%) | 17 (56.7%) | |||
| Family History | Yes | 9 (18.0%) | 8 (26.7%) | 0.84 | 0.359 |
| No | 41 (82.0%) | 22 (73.3%) | |||
Bisphenol was detected in serum using a Bisphenol A ELISA kit from MYBIOSOURCE with a detection range 200 ng/mL-3.12 ng/mL and sensitivity of Up to 0.6 ng/mL which was the cutoff used to define detection of bisphenol A in serum of subjects.
The percent of patients in the DCM group who had ECG abnormalities was 50%, while the percent of patients in the healthy control group who had ECG abnormalities was 10%, respectively (Table 2). There was a statistically significant moderate association between DCM and ECG abnormalities (X2= 13.2, P = 0.0003, V= 0.406).
Comparison between dilated cardiomyopathy and healthy controls groups regarding the ECG abnormalities.
| Dilated Cardiomyopathy Group No: 50 | Healthy Controls Group No: 30 | Chi-square test | P value | ||
|---|---|---|---|---|---|
| ECG Characteristics | Normal ECG | 25 (50.0%) | 27 (90.0%) | 13.2 | 0.0003 |
| Abnormal ECG including Atrial Fibrillation, Left Bundle Branch Block, and Right Bundle Branch Block | 25 (50.0%) | 3 (10.0%) |
The percent of patients in the DCM group who were exposed to BPA was 24%, while the percent of patients in the healthy control group who were exposed to BPA was 6.7%, respectively (Table 3). There was a statistically significant weak association between BPA exposure and DCM (X2= 3.9, P = 0.048, V = 0.221).
Comparison between dilated cardiomyopathy and healthy controls groups regarding Bisphenol A.
| Dilated Cardiomyopathy Group No: 50 | Healthy Controls Group No: 30 | Chi-square test | P value | |
|---|---|---|---|---|
| Bisphenol A detected | 12 (24.0%) | 2 (6.7%) | 3.9 | 0.048 |
| Bisphenol A not detected | 38 (76.0%) | 28 (93.3%) |
In the DCM group, the percent of patients who were New York Heart Association (NYHA) class II was 44%, the percent of patients who were NYHA class III was 34%, and the percent of patients who were NYHA class IV was 22%, respectively (Table 4). There was a non-significant association between BPA exposure and the NYHA classification of HF (X2= 2.5, P = 0.280).
Association between Bisphenol A and New York Heart Association classification of heart failure in patients with dilated cardiomyopathy
| Dilated Cardiomyopathy with detected Bisphenol A No: 12 | Dilated Cardiomyopathy with non-detected Bisphenol A No: 38 | Chi-square test | P value | |
|---|---|---|---|---|
| New York Heart Association Class II | 5 (41.7%) | 17 (44.7%) | 2.5 | 0.280 |
| New York Heart Association Class III | 6 (50.0%) | 11 (28.9%) | ||
| New York Heart Association Class IV | 1 (8.3%) | 10 (26.3%) |
In the DCM group, there was a statistically significant difference in the LA diameter of the study participants exposed to BPA versus not exposed to BPA (t(48) = 3.19, P = 0.050), a highly significant strong association between BPA exposure and RV dilatation (X2= 17.3, P = < 0.0001, V= 0.588), and a highly significant difference in the TAPSE of the study (t(48) = –3.16, P = < 0.0001) (Table 5).
Association between Bisphenol A and volumetric indexes of the left atrial diameter and right ventricular size and function in patients with dilated cardiomyopathy.
| Dilated Cardiomyopathy with detected Bisphenol A No: 12 | Dilated Cardiomyopathy with non-detected Bisphenol A No: 38 | Independent t-test | P value | ||
|---|---|---|---|---|---|
| Left Atrial Diameter Mean ± SD | 40.58 ± 4.34 | 37.39 ± 5.55 | 3.19 | 0.050 | |
| Dilated Cardiomyopathy with detected Bisphenol A | Dilated Cardiomyopathy with nondetected Bisphenol A | Chi-square test | P value | ||
| No: 12 | No: 38 | ||||
| Right Ventricle Size | Dilated Right Ventricle | 9 (75.0%) | 5 (13.2%) | 17.3 | < 0.0001 |
| Normal Right Ventricle | 3 (25.0%) | 33 (86.8%) | |||
| Dilated Cardiomyopathy with detected Bisphenol A | Dilated Cardiomyopathy with nondetected Bisphenol A | Independent t-test | P value | ||
| No: 12 | No: 38 | ||||
| Tricuspid Annular Plane Systolic Excursion Mean ± SD | 14.42 ± 1.93 | 17.58 ± 2.32 | –3.16 | < 0.0001 | |
Recent epidemiological studies have detected BPA in contaminated food and drinking water, wastewater, air and dust particles with an overall 90% prevalence rate of environmental and occupational exposure to BPA and its analogues12.
In 2012, the National Institute of Environmental Health Sciences and National Toxicology Program harmonized a consortium-based approach to address the gaps in research on the health effects of BPA13.
Three years later, Xiong et al investigated the association between BPA exposure and cardiovascular diseases.
They demonstrated increased serum levels of BPA in DCM patients compared with healthy controls.
Another study by Posnack et al. has concluded significant impairment of cardiac contractility in male and female rats when exposed to environmentally relevant doses of BPA14.
Our study compared the BPA serum levels between healthy controls and DCM patients. It included 80 patients in whom, 50 were DCM and 30 were controls. In our study, BPA was significantly more detected (P:0.04) among DCM patients than the healthy control group. Our study’s results were consistent with Xiong et al study.
As noted above, the percent of patients in the DCM group who were exposed to BPA was 24%, while the percent of patients in the healthy control group who were exposed to BPA was 6.7%, respectively, confirming a statistically significant association between BPA exposure and DCM.
The association between the impaired cardiac function in rats and exposure to BPA addressed in Posnack et al. animal study might explain our study’s findings of a significant decline in the RV function when DCM patients are exposed to BPA.
Correlation between laboratory and echocardiographic findings among DCM patients found that in participants exposed to BPA, the RV functions and dimensions were more affected than those not exposed to BPA with significant difference in TAPSE. (P = < 0.0001).
Also, the LA diameter showed statistically significant difference (p=0.050) between patients exposed versus non exposed to BPA. These findings highlight the generalized cardiac effect of BPA exposure not only on the LV.
The pathogenesis involves epigenetic changes, nuclear receptogeneralized cardiac in male anieexpexposure note to BPA for long time caused hypermethylation and decreased expression of Peroxisome proliferator activated receptor γ co-activator one alfa.
Treatment of pregnant female animal models with BPA from the 6th gestational day to just before delivery and from the first day after delivery to the end of the third week, caused inflammatory cellular infiltration and a focal fibrosis resulting in increased incidence and increased severity of cardiomyopathy16.
BPA was more detected in patients with DCM highlighting the possibility of an association between them. This rises the need of limitation of using plastics and other manufactured material from BPA and the use of other alternative to BPA in carbonated materials and plastic industries.
Our study was conducted on a well-balanced cohort with respect to baseline characteristics, risk factors, and NYHA class, with no missing data, enabling a robust per-protocol analysis. Echocardiographic outcomes were assessed and reported by investigators who were blinded to participant identity and laboratory results, thereby minimizing observer bias. To the best of our knowledge, the association between BPA exposure and volumetric indices of LA diameter as well as RV size and function in patients with DCM has not been previously investigated. Despite its strengths, the study limitations require consideration. It was a single centre cross-sectional study with a small sample size which didn’t allow us to investigate a causal relationship between the echocardiographic assessment outcomes and retrospective BPA exposure in patients with DCM vs healthy controls.
BPA exposure can vary depending on many factors including occupation, income and preferences which can affect how often people use plastic containers in cooking and storing food.
The prevalence of BPA exposure was more in patients with DCM vs healthy controls and the volumetric indexes of the LA diameter, and the RV size and function were worse in patients with DCM exposed to BPA versus patients with DCM not exposed to BPA.
What this study adds (what’s new):
- –
The association between BPA exposure and DCM is an observation which warrants further research as a case-control study to determine if there is a causal relationship between BPA exposure and DCM.
- –
Recommending use of BPA industrial alternatives in food and beverage containers.
- –
Recommending public campaigns to raise awareness about BPA environmental and health hazards.