Subscriber access provided by KEAN UNIV
Ecotoxicology and Human Environmental Health
Bisphenol A, chlorinated derivatives of Bisphenol A and occurrence of myocardial infarction in patients with type 2 diabetes: nested case-control studies in two European cohorts Chunyun Hu, Ben Schöttker, Nicolas Venisse, Frederike Limousi, Pierre Jean Saulnier, Marion Albouy-Llaty, Antoine DUPUIS, Hermann Brenner, Virginie Migeot, and Samy Hadjadj Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.9b02963 • Publication Date (Web): 16 Jul 2019 Downloaded from pubs.acs.org on July 18, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 24
Environmental Science & Technology
1
Bisphenol A, chlorinated derivatives of Bisphenol A and occurrence of myocardial infarction
2
in patients with type 2 diabetes: nested case-control studies in two European cohorts
3
Chunyun Hu *1,2,3, Ben Schöttker *4,5, Nicolas Venisse 3,6, Frédérike Limousi 1,2,3, Pierre Jean
4
Saulnier 2,3, Marion Albouy-Llaty 1,2,3, Antoine Dupuis 2, 3, 7, Hermann Brenner *4,5, Virginie
5
Migeot *1,2,3, Samy Hadjadj *2,3,8,9
6 7
1 CHU
Poitiers, Department of Public Health, F-86021 Poitiers, France
8
2 Université de Poitiers, School of Medicine and
9
3 INSERM
Pharmacy of Poitiers, F-86073 Poitiers, France
CIC 1402, CHU Poitiers, F-86021 Poitiers, France
Division of Clinical Epidemiology and Aging Research, Division of Preventive Oncology,
10
4
11
German Cancer Research Center (DKFZ), D-69120 Heidelberg, Germany
12
5
Network Aging Research, D-69115 Heidelberg, Germany
13
6
CHU Poitiers, Department of Toxicology and Pharmacokinetics, F-86021 Poitiers, France
14
7
CHU Poitiers, Department of Pharmacy, F-86021 Poitiers, France
15
8
CHU Poitiers, Department of Endocrinology, F-86021 Poitiers, France
16
9
L’institut du thorax, INSERM, CNRS, Univ Nantes, CHU Nantes, F-44093 Nantes, France
17 18
*These authors equally contributed to the present manuscript
19
Correspondence to : V Migeot :
[email protected] 20
CHU Poitiers, Department of Public Health, 2 rue de la Milétrie, F-86021 Poitiers, France
21 22
Key words: Bisphenol A – Chlorinated derivatives of bisphenol A- Myocardial infarction – Type 2
23
diabetes.
24
ACS Paragon Plus Environment
Environmental Science & Technology
25
ABSTRACT
26
A positive association between Bisphenol A (BPA) exposure and coronary heart disease has
27
been shown, but not in patients with type 2 diabetes (T2D). During the treatment of drinking
28
water, chlorination leads to the formation of chlorinated derivatives of Bisphenol A (ClxBPA),
29
that have higher estrogenic activity than BPA. No evidence exists for a relationship between
30
exposure to ClxBPA and myocardial infarction in patients with T2D. The objective of this study
31
was to evaluate the relationship between exposure to BPA, ClxBPA and the occurrence of
32
myocardial infarction (MI) in patients with T2D. Two nested case-control studies in two
33
independent European cohorts were performed. Each case with incident MI during follow-up
34
was matched to one control on age, sex and personal cardiovascular history in the same
35
cohort. Association between baseline urine concentrations of BPA and of ClxBPA and incident
36
MI was determined. Exposure to BPA was 31% in the ESTHER cohort and 18% in the
37
SURDIAGENE cohort. In a meta-analysis of the two studies, occurrence of MI was significantly
38
associated with urine BPA detection: adjusted OR=1.97 (1.05; 3.70), p=0.04. Exposure to
39
ClxBPA significantly differed in the SURDIAGENE and ESTHER studies: 24% and 8%, respectively
40
(p=0.0003). It was very strongly associated with MI in the SURDIAGENE cohort with an
41
adjusted odds ratio (OR) of 14.15 (95% confidence interval: 2.77 - 72.40) but this association
42
was not replicated in the ESTHER study: adjusted OR: 0.17 (0.02 to 1.23). Whether these
43
results may be explained by different water chlorination processes in France and Germany,
44
resulting in different ClxBPA exposure levels, requires further investigation.
45
ACS Paragon Plus Environment
Page 2 of 24
Page 3 of 24
Environmental Science & Technology
46
INTRODUCTION
47
People with diabetes mellitus (DM) represent a large and growing population worldwide. It has been
48
shown that diabetes increases the risk of coronary heart disease (CHD) 1 and mortality after a first
49
myocardial infarction (MI) 2. People with type 2 diabetes (T2D) are prone to cardiovascular disease
50
(CVD), including MI, and the risk of CHD mortality has been suggested to be equivalent in patients with
51
T2D with no history of MI to the risk in non-diabetes persons with a history of MI 3. The possible
52
contribution to this increased risk of environmental factors such as endocrine disruptors has yet to be
53
established. Patients with T2D risk can be seen as a group of peculiar interest for the identification of
54
environmental factors related to cardiovascular complications.
55 56
Bisphenol A (BPA) is a synthetic monomer considered as an endocrine-disrupting chemical inasmuch
57
as it exhibits estrogenic activity 4. The general population in Western countries is exposed to and/or at
58
risk from exposure to BPA 5. Exposure is ubiquitous in daily life, mainly through packaged food and
59
beverages contaminated by BPA and migrating from the packaging material 67. During the treatment
60
of drinking water, chlorination can easily lead to the formation of chlorinated derivatives of BPA,
61
resulting in production of monochloro-bisphenol A (MCBPA), dichloro-bisphenol A (DCBPA), trichloro-
62
bisphenol A (TCBPA) and tetrachloro-bisphenol A (TTCBPA). Human exposure to chlorinated
63
derivatives of BPA mainly originates from drinking water, except for TTCBPA, which is used as a flame-
64
retardant 8. Several studies have demonstrated, in vitro or in vivo, higher estrogenic activity with
65
chlorinated derivatives than with BPA, especially MCBPA and DCBPA
66
proliferator-activated receptor gamma (PPAR) agonist effect
67
separate environmental sources and could translate into different health risks. Evidence for the
68
association of BPA with T2D has been inconsistent in human studies 12 13 14 15. Notwithstanding a recent
69
meta-analysis suggesting a positive association between urinary concentration of BPA and CVD/CHD
70
16,
71
particularly in patients with T2D. Recently, Andra showed a weak positive association between urinary
11.
9 10,
and also a peroxisome
ClxBPA and BPA probably reflect
we still lack consolidated clinical evidence of the association between BPA exposure and CVD/CHD,
ACS Paragon Plus Environment
Environmental Science & Technology
72
mono-chloro-BPA and obesity 17 or with T2D 18. To our knowledge, no previous study has investigated
73
the relationship between exposure to ClxBPA and CVD/CHD, either in the general population or in T2D
74
patients.
75 76
We consequently evaluated the relationship between exposure to BPA and ClxBPA and the risk of
77
myocardial infarction SURDIAGENE cohort
78
German ESTHER cohort as a replication step 20.
19
as a discovery step, and in T2D participants from the
79 80
METHODS
81
Design: We performed a nested case-control study in the SURDIAGENE cohort (discovery step) and in
82
the ESTHER cohort (replication step).
83 84
Settings and participants:
85
SURDIAGENE discovery cohort: The SURDIAGENE study is a French single-center inception cohort of
86
T2D patients regularly visiting the diabetes department at Poitiers University Hospital, France (23),
87
including 1,468 participants. Patients were enrolled from 2002 to 2012. Main recruiting criteria were
88
type 2 diabetes, registration to the French social security and no evidence of non-diabetic renal
89
disease. The current report considered adjudicated outcomes, with median follow-up time for the
90
whole cohort of 6.4 years, as previously published 19. Study outcome updates were performed every 2
91
years since 2007. The data from 2013 were considered for the current analysis. Vital status was
92
ascertained using the French death registry (CepiDC). The cause of death was specified according to
93
hospital discharge information and/or interview of the GP and/or interrogation of the CepIDC. Fatal
94
MI was considered for death coded with ICD-10 codes I21-I23. The Poitiers University Hospital Ethics
95
Committee approved the design (CPP Ouest III). All participants in the study gave their informed
96
written consent 19 and the study was conducted in accordance with the Declaration of Helsinki.
97
ACS Paragon Plus Environment
Page 4 of 24
Page 5 of 24
Environmental Science & Technology
98
ESTHER replication cohort: ESTHER is an ongoing population-based cohort study including 9,949 older
99
adults resident in the state of Saarland, Germany, described in detail elsewhere 21. In brief, participants
100
were recruited between July 2000 and December 2002 by their general practitioner (GP) during a
101
regular health check-up. Inclusion criteria were residence in the state of Saarland, age between 50 and
102
74 years, and sufficient knowledge of the German language. The current report focuses on 1,375
103
ESTHER participants with physician-diagnosed T2D at baseline 20. Follow-up questionnaires, including
104
information on incident MI, were sent to study participants and their GPs after 2, 5 and 8 years with
105
response rates among survivors of 96%, 87% and 79%, respectively. Fatal MI events between 2000 and
106
2010 were identified by inquiry at the residents’ registration offices in Saarland and information on
107
the vital status could be obtained for 99.9% of the ESTHER participants. Death certificates were
108
provided for 97.7% of deceased participants by public health departments. Fatal MI events were
109
identified by deaths coded with ICD-10 codes I21-I23. The ESTHER study was approved by the Ethics
110
Committees of the Medical Faculty, University of Heidelberg and by the Medical Association of
111
Saarland and was conducted in accordance with the Declaration of Helsinki. Written informed consent
112
was obtained from all participants.
113
Sample size calculation: For the discovery cohort all patients from the Surdiagene study presenting a
114
myocardial infarction were included, as cases. Based on the study from Liao and Kannan 22, exposure
115
to Clx-BPA was estimated to be 20%. Then, using one control for each case enables to detect, at 95%
116
probability, an association between exposure to Clx-BPA and MI with an estimated Odds Ratio equal
117
to 2.53. Due to high cost of the assay, increasing the control-to-case ratio was not possible. Similar
118
hypotheses were used for sample size calculation for the replication cohort.
119
Participants: In the SURDIAGENE cohort, cases were defined as all eligible participants with T2D who
120
developed myocardial infarction during the follow-up period (2002-2013). MI was defined according
121
to the universal definition 23 with adjudication by an independent clinical outcome committee in the
122
SURDIAGENE cohort (see composition in appendix). In the ESTHER cohort, self-reported MI was only
123
considered as a case if it was confirmed by medical records from the study participants' GP. In addition,
ACS Paragon Plus Environment
Environmental Science & Technology
124
MIs recorded in death certificates were considered as cases. Of note, we considered both first-
125
occurring and recurrent MI during follow-up for the SURDIAGENE participants, while only first-
126
occurring MI was used for the ESTHER participants.
127 128
Controls were selected by incidence density sampling, which involved matching each case to a sample
129
of those who were at risk at the time of case occurrence 24. We chose one control with T2D for each
130
case, matched on age (± 3 years), sex, and personal history of cardiovascular disease at study entry
131
(stroke or MI for the SURDIAGENE and stroke for the ESTHER study participants). The main exclusion
132
criteria for both nested case-control studies were: no available baseline urine sample and estimated
133
glomerular filtration rate (eGFR) < 30 ml/min/1.73m2). In addition, in the ESTHER study, subjects with
134
a history of MI before baseline were excluded. After inclusion of cases and controls, some urine
135
samples were unfortunately damaged leading to the exclusion of 11 case-control pairs in the ESTHER
136
study. The baseline spot urine samples of all the eligible participants of the two studies were collected
137
and stored at -80°C. The SURDIAGENE samples were stored and handled at the CHU Poitiers biobank
138
(CRB BB0033-00068). We collected the main confounding factors at baseline that may influence the
139
occurrence of MI. Variables included were smoking (from self-reported status as non-smoker vs.
140
current or former smoker), body mass index, systolic and diastolic pressure, total plasma cholesterol,
141
LDL-cholesterol, HDL-cholesterol, Triglycerids, HbA1C, and glomerular filtration rate (MDRD).
142
Assessment of BPA and ClxBPA exposure: Three hundred microliters of urine sample were used for
143
analysis, after storage at -80°C at CRB, CHU de Poitiers: Biobanking facility BB0033-00068. Urine
144
concentrations of unconjugated BPA, MCBPA, DCBPA, TCBPA and TTCBPA were measured by the same
145
laboratory in Poitiers University Hospital using Liquid Chromatography-tandem Mass Spectrometry
146
technology. Technicians were blinded for the case/control status. The limits of detection (LOD) were
147
0.25 and 0.025 ng/mL for BPA and ClxBPA, respectively. The limits of quantification (LOQ) were 0.5 and
148
0.05 ng/mL for BPA and ClxBPA, respectively 25. It means that when above the LOQ (0.05 ng/mL), the
ACS Paragon Plus Environment
Page 6 of 24
Page 7 of 24
Environmental Science & Technology
149
concentration can be determined with adequate accuracy and precision while when below the LOQ
150
(0.025 ng/mL), one can guarantee with a 95% level of confidence
151
that the compound is present but its concentration cannot be determined with adequate accuracy and
152
precision. Otherwise when below the LOD (LOD (%)
P
BPA
30 (18.1%)
19 (22.9%)
11 (13.2%)
0.15
CBPA
18 (10.8%)
16 (19.3%)
2 (2.4%)
0.0007
DCBPA
20 (12.0%)
16 (19.3%)
4 (4.8%)
0.007
TCBPA
24 (14.5%)
20 (24.1%)
4 (4.8%)
0.0006
TTCBPA
33 (19.9%)
30 (36.1%)
3 (3.6%)