Article
Subscriber access provided by SUNY DOWNSTATE
Dietary and household sources of prenatal exposure to polybrominated diphenyl ethers (PBDEs) in the INMA birth cohort (Spain) Olga Costa, Maria Jose Lopez-Espinosa, Esther Vizcaino , Mario Murcia, Carmen Iñiguez, Eva María Navarrete-Muñoz, Joan O. Grimalt, Adonina Tardon, Ferran Ballester, and Ana Fernandez-Somoano Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.5b06263 • Publication Date (Web): 28 Apr 2016 Downloaded from http://pubs.acs.org on May 3, 2016
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 free 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 accessible to all readers and 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.
Environmental Science & Technology 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 35
Environmental Science & Technology
1
Dietary and household sources of prenatal exposure to polybrominated diphenyl
2
ethers (PBDEs) in the INMA birth cohort (Spain)
3
Olga Costaa, Maria-Jose Lopez-Espinosaa,b*, Esther Vizcainoc,d, Mario Murciab,a,
4
Carmen Iñigueza,b, Eva M. Navarrete-Muñozb,e, Joan O. Grimaltd, Adonina Tardonc,b,
5
Ferran Ballestera,b, Ana Fernandez-Somoanoc,b
6 7
a
8
Jaume I–Universitat de València, Avenida de Catalunya 21, 46020 Valencia, Spain
9
b
Epidemiology and Environmental Health Joint Research Unit, FISABIO–Universitat
Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP),
10
Calle Monforte de Lemos 3-5, 28029 Madrid, Spain.
11
c
12
del Cristo s/n, 33006 Oviedo, Asturias, Spain.
13
d
14
Water Research (IDÆA-CSIC), Jordi Girona 18, 08034 Barcelona, Spain.
15
e
16
Juan de Alicante, 03550 Alicante, Spain.
Department of Preventive Medicine and Public Health, University of Oviedo, Campus
Department of Environmental Chemistry, Institute of Environmental Assessment and
Department of Public Health, Miguel Hernandez University, Ctra. Nacional 332, San
17 18
Corresponding author: *Maria-Jose Lopez-Espinosa
19
Foundation for the Promotion of Health and Biomedical Research in the Valencian
20
Region, FISABIO-Public Health, Valencia, Spain, Avda Catalunya 21, 46020 Valencia,
21
Spain. E-mail:
[email protected]; Phone: (+34) 961925941.
22
Conflict of interest: The authors declare no competing financial interest.
23
Word count: 7631
24
Full text: 4931
25
Tables and figures: 2700
26 ACS Paragon Plus Environment
1
Environmental Science & Technology
Page 2 of 35
27
Abstract
28
This study looked at predictors of exposure to polybrominated diphenyl ethers (PBDEs)
29
with a focus on dietary and household-level factors. Concentrations of BDE-47, -99,
30
-153, and -209 and their sum (∑PBDEs) were measured in cord serum. Spanish women
31
(n=541) completed two semi-quantitative food frequency questionnaires during the first
32
and third trimesters of pregnancy. The daily mean intake (grams) of eggs, dairy
33
products, meat, cereals and pasta, vegetables and pulses, fruits, shellfish and
34
cephalopods, and fish, and the weekly mean intake (servings) of lean, large oily, other
35
oily, and other fish from both questionnaires were averaged. Information on house size,
36
curtains and carpets at home, mattress type, housekeeping frequency, and television use
37
was also collected later in gestation. Multivariate censored regression was used to assess
38
the association between PBDE concentration (log2 transformed) and potential
39
predictors. BDE-47, -99, -209 and ∑PBDE concentrations increased by 13.6%(95%
40
CI:0.0, 29.0%), 21.1%(2.3, 43.5%), 21.7%(0.4, 47.5%) and 11.5%(2.2, 21.7%),
41
respectively, per interquartile range increment in daily intake of shellfish and
42
cephalopods. Fish intake was associated with BDE-99 (20.8%[1.7, 43.4%]). When fish
43
was disaggregated by types, BDE-99 and ∑PBDEs increased by 13.8%(4.0, 24.7%) and
44
5.7%(0.8, 10.8%), respectively, per 1-serving/week increment in large oily fish intake.
45
BDE-153 was associated with higher housekeeping frequency (35.9%[0.4, 83.9%]) and
46
BDE-209 with foam mattress use (48.9%[5.8, 109.7%]). In conclusion, seafood
47
consumption, higher housekeeping frequency, and foam mattress were associated with
48
prenatal PBDE exposure.
49
Key words: Cord, diet, fish, household sources, PBDEs, prenatal exposure.
50 51
ACS Paragon Plus Environment
2
Page 3 of 35
Environmental Science & Technology
52
TOC Art:
53 54
ACS Paragon Plus Environment
3
Environmental Science & Technology
Page 4 of 35
55
INTRODUCTION
56
Polybrominated diphenyl ethers (PBDEs) are synthetic chemicals commonly used to
57
reduce flammability in many types of commercial and household products, such as
58
plastics, electronics, furniture, textiles, and construction materials.(1) The use and
59
production of DecaBDE has been phased out in the US and restricted in Europe, but it is
60
still produced in other parts of the world.(2,
61
mixtures have been banned in many parts of the world(4, 5) and they were added to the
62
list of persistent organic pollutants under the Stockholm Convention in 2009.(6) PBDEs
63
are not chemically bound to the products that contain them, which makes these
64
contaminants more likely to leach out from the goods into house dust, which is
65
unintentionally inhaled and ingested.(1) Hence, household dust may make a considerable
66
contribution to human PBDE body burden.(7-9)
67
Nowadays, the in-use reservoirs of PBDEs, such as electronics, plastics, or household
68
furnishing, are considerable. In addition, disposal reservoirs of goods containing PBDEs
69
(e.g., landfills, sewage and hazardous waste treatment plants, and e-waste recycling
70
facilities) may also be important for the dispersal of PBDEs and their incorporation and
71
biomagnification along the aquatic food chain.(10) Therefore, another important source
72
of PBDE exposure for non-occupationally exposed populations is diet,(7, 8) since these
73
chemicals are persistent, lipophilic, accumulate in organisms, and biomagnify along the
74
food chain.(11) Studies of PBDE concentrations in foodstuffs have highlighted seafood
75
in Asia and Europe and meat and meat products in the US as the main contributors to
76
daily food intake of these contaminants.(7, 8)
77
PBDEs have been detected in the blood of pregnant women and children(12-14) and
78
concerns have recently emerged, given their endocrine-disrupting properties.(15) In
79
humans, the fetal period is one of the stages of greatest vulnerability to PBDE exposure,
3)
The PentaBDE and OctaBDE technical
ACS Paragon Plus Environment
4
Page 5 of 35
Environmental Science & Technology
80
since even low levels of synthetic compounds could produce irreversible and irreparable
81
damage.(15) While few human studies on the possible effects of PBDE exposure on fetal
82
development have been published to date, recent evidence suggests that prenatal
83
exposure to these compounds could be associated with disturbed hormone levels during
84
pregnancy,(16) fetal growth,(17) and neurodevelopment in infants and toddlers.(18, 19) It is
85
thus essential to identify and describe sources of prenatal exposure to PBDEs. To date,
86
sociodemographic predictors of the body burden of these compounds have been studied
87
quite extensively,(20-25) but data on food intake and indoor exposure are scarce.(26, 27) The
88
present study aimed to look at predictors of cord serum concentrations of four PBDEs
89
(BDE-47, -99, -153, and -209) and their sum (∑PBDEs) with a focus on dietary and
90
household-level factors.
91 92
EXPERIMENTAL SECTION
93
Study design
94
Population-based birth cohorts were established in several areas of Spain as part of the
95
INMA Project.(28) The population in the present study consisted of pregnant women
96
who agreed to participate in the INMA cohorts of Asturias and Valencia, which are the
97
only two INMA cohorts with PBDEs measured in cord serum. The Asturias cohort is
98
located in northern Spain on the shores of the “Cantabrian Sea” (in the Atlantic Ocean)
99
and the Valencia cohort is located in eastern Spain on the Mediterranean coast.
100
A total of 1,349 eligible women (age ≥ 16 years, singleton pregnancy, enrollment at 10–
101
13 weeks of gestation, no assisted conception, delivery scheduled at the reference
102
hospital, and no communication handicap) agreed to participate (recruitment period:
103
2003-2007). The participation rates were 45% and 54% in the Asturias and Valencia
104
cohorts, respectively. Excluding the women who withdrew from the study, were lost to
105
follow-up, or with induced or spontaneous abortions or fetal deaths, 1,272 (94%)
ACS Paragon Plus Environment
5
Environmental Science & Technology
Page 6 of 35
106
women were followed up to delivery (2004-2008). Cord blood samples were collected
107
from 853 (67%) of these participants. A total of 541 cord serum samples (around 99%
108
[n=322] and 42% [n=219] from the Asturias and Valencia cohorts, respectively) were
109
available for PBDE determinations and were included in this study. In Valencia, the
110
lower proportion of existing samples for PBDE determinations was due to logistic
111
reasons, since only a sub-sample of all the initially collected cord blood samples were
112
available for the determination of these contaminants.
113
All the mothers gave their written informed consent prior to inclusion. The Ethics
114
Committees of the San Agustin Hospital in Aviles (Asturias) and La Fe Hospital in
115
Valencia approved the research protocol.
116
Serum collection
117
Cord blood was collected at birth by trained midwives following the same protocol in
118
both birth cohorts.(28) Briefly, venous cord blood (volume: 5 mL) was collected with a
119
glass syringe in the delivery room within 10 minutes after placental delivery. Part of the
120
collected samples were placed in Vacutainer blood collection tubes and stored in the
121
dark at 4ºC (1-2 hours from the extraction) until centrifugation for 15 minutes at 2500-
122
3000 rpm. The serum obtained was aliquoted in 2 mL glass criyotubes and
123
anonymously stored at -80°C.
124
Analysis of PBDEs
125
The analytical methods and quality control procedures employed in the laboratory have
126
been described elsewhere.(29) Briefly, 1 mL of serum was spiked with the surrogate
127
standards tetrabromobenzene (TBB) and decachlorobiphenyl (PCB 209) and vortex
128
stirred for 30 s at 2,000 rpm. n-Hexane (3 mL) was added, followed by concentrated
129
sulfuric acid (2 mL). After reacting, the mixture was stirred for 30 s and the supernatant
130
n-hexane phase was separated by centrifugation. The remaining sulfuric acid solution
ACS Paragon Plus Environment
6
Page 7 of 35
Environmental Science & Technology
131
was re-extracted twice with 2 mL of n-hexane (each by 30 s of stirring and
132
centrifugation). The combined n-hexane extracts (7 mL) were additionally cleaned with
133
sulfuric acid (2 mL, stirring for 30 s). Then, the n-hexane phase was separated by
134
centrifugation and reduced to a small volume under a gentle nitrogen stream. The
135
extract was transferred to gas chromatography (GC) vials using four 25 µL isooctane
136
rinses. PCB 142, BDE-118 (20 µl) and [13C]-BDE-209 (10 µL) were added as internal
137
standards before injection. BDE congeners were analyzed by GC coupled to mass
138
spectrometry in chemical ionization and negative ion recording.
139
A total of 14 PBDEs were analyzed. BDE-17, BDE-66, BDE-71, BDE-85 BDE-138 and
140
BDE-190 were detected in limit of detection [LOD]) are also explored since available data on
145
dietary and household-level factors of pre- or postnatal BDE-209 exposure are scarce.
146
In addition, this congener is still in use in some parts of the world, including Europe,(3)
147
and, consequently, concentrations in human bodies are very likely to increase. LODs
148
were calculated as three times the standard deviation (SD) of the procedural blank
149
levels. The LODs for BDE-47, -99, -153, and -209 were 2.59, 2.38, 1.24, and 9.12
150
pg/mL in Asturias, and 2.30, 2.10, 1.19, and 5.94 pg/mL in Valencia. Samples were
151
analyzed in two different batches, one per cohort, and therefore LODs were somewhat
152
different for each cohort. The laboratory is in compliance with the Arctic Monitoring
153
and Assessment Program (AMAP) for persistent organic pollutants in human serum
154
(Centre de Toxicologie, Institut National de Sante Publique du Quebec). The same
ACS Paragon Plus Environment
7
Environmental Science & Technology
Page 8 of 35
155
samples from the AMAP were analyzed in the two different batches to ensure
156
comparable results and concentrations were always within 20% of the consensus values.
157
Enzymatic techniques were used to determine total cholesterol and triglycerides, and
158
total serum lipid concentrations were calculated from these measurements.(30)
159
Dietary and household sources of exposure
160
Two semi-quantitative food frequency questionnaires (FFQs) were administered by
161
trained interviewers in two different interviews: the first one (10−13 weeks) collected
162
information from the last menstrual period to the end of the first trimester, and the
163
second FFQ (28−32 weeks) covered the period since the first FFQ was administered.
164
The 101-item FFQ was an adapted version of Willett’s questionnaire,(31) developed and
165
validated for use among adults living in Spain(32) and later in pregnant Mediterranean
166
women, with satisfactory coefficients for validity and reproducibility.(33) Participants
167
were asked how often, on average, they had consumed each item type over the past 3−4
168
months (with nine possible responses ranging from “never or less than once a month” to
169
“six or more per day”). Their responses were transformed to average daily intake in
170
grams for each food item using standard units or serving sizes specified in the FFQ.
171
Each food group (eggs, dairy products, meat, cereals and pasta, vegetables and pulses,
172
fruits, shellfish and cephalopods, and fish) was obtained by summing the daily intake of
173
individual items included in the group. The fish group was disaggregated into lean fish,
174
large oily fish, other oily fish, and other fish. The items included in these groups are
175
described in the Supporting Information (Table S1). In addition, the total energy intake
176
was primarily obtained from food composition tables of the US Department of
177
Agriculture(34) and other tables published for Spanish foods.(35)
178
In the present study, the mean of the two FFQs for each item was used as an estimation
179
of daily average intake of each food item. For statistical analysis, the intake of food
ACS Paragon Plus Environment
8
Page 9 of 35
Environmental Science & Technology
180
groups was expressed in daily grams, and the intakes of specific fish groups were
181
expressed in weekly servings.
182
Questions that may be related with PBDE exposure were also selected from a
183
questionnaire administered at gestational weeks 28−32 on sociodemographic and
184
lifestyle variables(1): house size (m2), curtains at home (yes and no), carpets at home
185
(yes and no), type of mattress (foam: mattress with one or more types of foam
186
[polyurethane, viscoelastic, and latex, among others] as the support system, and
187
innerspring: mattress with a steel coil support system covered by padding or upholstery
188
materials, which can include various foams, fiber, and additional layers of smaller steel
189
springs), housekeeping frequency (≤ or >1 times/week, including sweeping, vacuuming
190
dusting, and mopping), and television use (hours/week). Questions on household-level
191
factors of PBDE exposure were only requested later in gestation and referred to the
192
whole pregnancy.
193
Covariates
194
The pregnant women completed two detailed in-person questionnaires (weeks 10−13,
195
and 28−32) on anthropometric, sociodemographic and lifestyle information. The
196
maternal variables considered to be included in the models were: age (years), pre-
197
pregnancy body mass index (BMI, kg/m2), country of origin (Spain and other), zone of
198
residence (rural and non-rural), working at week 12 of pregnancy (yes and no), parity
199
(primiparous and multiparous), previous lactation (none, and < or ≥6 months, since
200
lactation is a mechanism for the excretion of PBDEs(36)), education (until primary,
201
secondary, and university studies), social class (Class I: managerial jobs, senior
202
technical staff, and commercial managers, Class II: skilled non-manual workers, Class
203
III: manual workers) defined according to the most privileged occupation of the mother
204
or the father during pregnancy using a widely used Spanish adaptation of the
ACS Paragon Plus Environment
9
Environmental Science & Technology
Page 10 of 35
205
international ISCO88 coding system,(37) smoking at week 12 of pregnancy (yes and no),
206
passive smoking (yes and no) in at least two environments from among the home,
207
workplace and leisure areas/restaurants, alcohol intake at week 12 of pregnancy (yes
208
and no), and season of the last menstrual period.
209
Data analyses
210
For descriptive purposes, a number (percentage) is presented for categorical variables,
211
and mean and SD are given for continuous variables. Differences between the two
212
cohorts were explored by Chi square, t-Student, or Mann-Whitney tests, while logistic
213
regression models adjusted by cohort were used to explore the differences in study
214
variables between the excluded and included women. Percentage above LOD, geometric
215
mean (GM), geometric SD (GSD), median and percentiles (P) 25, 75 and 95 of PBDEs
216
are presented. Contaminant concentrations are expressed in pg/mL or ng/g lipid, the
217
latter dividing serum residue concentrations by total serum lipid levels. Missing values
218
of lipids (n=48) were imputed at random based on a normal probability distribution
219
using the mean and the SD of the lipids as parameters. Censored linear regression
220
models were used to estimate the GM and GSD of the PBDEs and to explore the
221
differences in contaminant concentrations between cohorts.
222
For statistical purposes, censored linear regression models were also performed to study
223
the association between PBDEs (after log2-transformation [log2(PBDE)] to account for
224
right-skewed distributions) and sources of exposure. When there are values below LOD
225
in the dependent variable, censored models provide maximum likelihood estimates of
226
the coefficients.(38) A two-step procedure was used for model building. First, covariates
227
associated with each PBDE at the p-value
