Prenatal exposure to organochlorine compounds and ...



Prenatal exposure to polybrominated flame retardants and fetal growth

Maria-Jose Lopez-Espinosa1,2*, Olga Costa1,2*, Esther Vizcaino3,4,5, Mario Murcia2,1, Ana Fernandez-Somoano2,3, Carmen Iñiguez1,2, Sabrina Llop1,2, Joan O. Grimalt4, Ferran Ballester1,2, Adonina Tardon3,2

*Equal contribution

1FISABIO–Universitat de València–Universitat Jaume I Joint Research Unit, 46020, Valencia, Spain, Maria-Jose Lopez-Espinosa Senior Research Fellow, Olga Costa PhD Student, Mario Murcia PhD Student, Carmen Iñiguez Senior Research Fellow, Sabrina Llop Senior Research Fellow, Ferran Ballester Senior Lecturer. 2Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP), Spain, Maria-Jose Lopez-Espinosa Senior Research Fellow, Olga Costa PhD Student, Mario Murcia PhD Student, Ana Fernandez-Somoano Research Fellow, Carmen Iñiguez Senior Research Fellow, Sabrina Llop Senior Research Fellow, Ferran Ballester Senior Lecturer, Adonina Tardon Senior Lecturer. 3Department of Preventive Medicine and Public Health, University of Oviedo, 33006, Oviedo, Asturias Spain, Esther Vizcaino Research Fellow, Ana Fernandez-Somoano Research Fellow, Adonina Tardon Senior Lecturer. 4Department of Environmental Chemistry, Institute of Environmental Assessment and Water Research (IDÆA-CSIC), 08034, Barcelona, Spain, Esther Vizcaino Research Fellow, Joan O. Grimalt Professorial Fellow. 5Department of Surgery and Cancer, Imperial College of London, SW7 2AZ, London, United Kingdom, Esther Vizcaino Research Fellow.

Corresponding author: Maria-Jose Lopez-Espinosa

E-mail: lopez_josesp@gva.es; Phone: (+34) 96 1925943; Fax: (+34) 96 1925703

Short running head: Polybrominated flame retardants and fetal growth

Key words: Fetal growth, ultrasound measures, maternal serum, cord serum, PBDE, prenatal exposure.

ABSTRACT

Objectives To investigate the relation between PBDEs and fetal growth or newborn anthropometry.

Design Birth cohort study established in 2004-2007.

Setting Asturias and Valencia, Spain.

Participants 686 mothers and their newborns

Main outcome measure PBDE congeners (BDE-47, -99, -153, -154 and -209) were determined in the serum of 670 mothers at 12 weeks of pregnancy and 534 samples of umbilical cord serum. Estimated fetal weight (EFW), abdominal circumference (AC), femur length (FL), and biparietal diameter (BPD) during gestation as well as birth weight (BW), birth length (BL), and head circumference (HC) at delivery were all measured. Growth at 12-20 and 20-34 gestational weeks and newborn anthropometry were both assessed by standard deviation (SD) scores adjusted for constitutional characteristics. The relation between PBDE congeners, their sum (ΣPBDEs) and outcomes was studied by multiple linear regression.

Results An inverse association was found between ΣPBDEs and AC, EFW, and BPD at 20-34 weeks and HC at birth. Regarding congeners, the association was clearer with BDE-99; an inverse association was found with AC, EFW, and BPD at 20-34 weeks, and BW and HC at delivery. Concerning matrices, more robust associations with head size were found when using maternal serum, while the effects on the rest of the parameters were clearer with cord serum. The range of decrease was 1.3-3.5% for each 2-fold increase in PBDE concentrations.

Conclusion PBDEs may impair fetal growth in late pregnancy and be observable at birth.

What is already know on this subject

PBDE human exposure has recently raised concern due to the endocrine disrupting properties of these substances. The fetal period is one of the most vulnerable stages to the exposure to these chemicals. However, the few studies on PBDE exposure and birth size have failed to yield conclusive results, and no reports have dealt with its possible effects on fetal growth during pregnancy.

What this study adds

This is the first study showing the specific body segments that may be affected by PBDEs during pregnancy and the largest study on newborn size conducted to date. Our results showed a decrease in weight, abdominal circumference and head diameter in late pregnancy associated with PBDEs and these effects were also visible at birth.

INTRODUCTION

Polybrominated diphenyl ethers (PBDEs) are flame retardants commonly used in many types of household and commercial products.1 Humans are exposed to these contaminants via several sources, including diet and indoor environment.2;3 PBDEs can cross the placenta and have been found in cord samples of different birth cohorts worldwide.4

Even though PBDEs have been widely used for several decades and humans are exposed to them on a daily basis, little is known about their possible effects on development. Thus, epidemiological reports on fetal growth are scarce, have chosen anthropometrical size at birth as a proxy measure of in utero development, and have failed to yield conclusive results. Some of these studies reported an inverse association with birth weight5-8 or length,9 whereas others found no association with anthropometric measures at delivery.10;11

Fetal growth is a good determinant of perinatal and postnatal health. In fact, the ‘fetal programming’ hypothesis proposed by Barker12 suggests that impaired fetal growth leads to small but permanent deficits in childhood development and is also a risk factor of some chronic diseases in adulthood, such as diabetes, hypertension, and heart disease.13 Consequently, the possible impact of PBDEs on fetal growth is a matter of public concern and more studies are warranted.

The main aim of the present study is to investigate the relation between maternal and cord concentrations of five PBDE congeners (BDE-47, -99, -153, -154 and -209) and their sum (ΣPBDEs) and growth at 12-20 and 20-34 weeks of pregnancy by using longitudinal fetal ultrasonic measurements and anthropometric measures at birth. This work was undertaken within the INMA (INfancia y Medio Ambiente – Environment and Childhood) Project from Spain.

MATERIAL AND METHODS

Study design and population

The INMA Project is a mother-and-child cohort study established in different areas of Spain following a common protocol.14 This study included the INMA cohorts of Asturias and Valencia since serum samples were available for PBDE determinations. The Hospital Ethics Committees of each region approved the research protocol, and all mothers gave their written informed consent between November 2003 and June 2007, prior to inclusion.

A total of 1349 eligible women (≥16 years, singleton pregnancy, enrolment at 10-13 weeks of gestation, non-assisted conception, delivery scheduled at the reference hospital, and no communication handicap) were recruited in the first trimester of pregnancy. Excluding the women who withdrew from the study, were lost to follow-up, and with induced or spontaneous abortions or fetal deaths, 1272 (94%) women were followed up to delivery.

In the present study, the sample size was 686 mothers and their newborns (Table 1) with at least two valid ultrasounds and samples available for PBDE determinations in maternal (n=670) and/or umbilical cord (n=534) serum. Around 98% of the women from the Asturias cohort were included in the present study and population characteristics did not differ between included and excluded women. A total of 28% of the women from Valencia were selected at random and included in the present study. The lower proportion of Valencia women in the study is due to the lack of serum samples available for PBDE determinations. There were no differences in population characteristics between them either, except in the case of a lower proportion of included mothers with non-rural residence (3%) and not working during pregnancy (12%) compared to non-included women (7% and 19%, respectively, data not shown).

PBDE exposure assessment

The analytical methods and quality control procedures used in the laboratory have been described elsewhere.15;16 Briefly, concentrations of 14 PBDEs were analyzed using gas chromatography coupled to a mass spectrometer (Detection limit (LOD) range: 0.0006-0.006 ng/mL). In this study, results of those PBDEs with maternal or/and cord concentrations with a detection frequency >50% (BDE-47, -99, -153, -154 and -209) were presented. The laboratory complies with the Arctic Monitoring and Assessment Program (AMAP) for persistent organic pollutants in human serum (Centre de Toxicologie, Institut National de Santé Publique du Québec).

Enzymatic techniques were used to determine total cholesterol and triglycerides, and total serum lipid concentrations were calculated as described by Phillips.17 Means (standard deviations [SDs]) of total lipid contents in maternal (n=473) and cord (n=486) serum were 5.8 (1.26) and 2.6 (0.51) mg/mL, respectively. We used individual total lipid values to calculate PBDE concentrations on a lipid content basis.

Fetal outcomes

Specialized obstetricians performed ultrasound examinations in routinely scheduled antenatal care visits in gestational weeks 12, 20 and 34. We also had access to any other ultrasound scans performed during pregnancy at the same hospital. From 2 to 8 valid ultrasound measurements were obtained per subject between the 7th and 41st weeks of gestation. Gestational age was based on the self-reported date of the last menstrual period, but an estimation based on an early crown–rump length measurement was considered if the self-reported and estimated dates differed by ≥7 days.18 Women with a difference of ≥3 weeks were excluded from the study (n=8). Data outside the range of the mean ±4 SDs for each gestational age were also eliminated to avoid the influence of extreme values (n≤5 for all parameters).

Fetal measures were abdominal circumference (AC), biparietal diameter (BPD) and femur length (FL). Additionally, estimated fetal weight (EFW) was calculated using the Hadlock algorithm.19 Linear-mixed models20 were used separately in each cohort to obtain longitudinal growth curves for each outcome. Models were adjusted for constitutional factors known to affect fetal growth (maternal height, age, parity, country of origin and pre-pregnancy weight, father’s height and fetal gender).

Fetal curves were used to calculate conditional SD scores for 12-20 and 20-32 weeks of gestation, which describe the growth of a fetus during these time intervals, i.e., the size at the final time point is evaluated using conditional mean and SD values based on the size at the initial time point. Information on fetal growth modeling has been reported previously.21

Newborn outcomes

Newborns were weighed (g) at birth by the midwife in the delivery room, and the length (cm) and head circumference (cm) were measured within the first 12 hours of life by a nurse in the hospital ward. Outcome variables were gestational age SD scores for birth weight (BW), birth length (BL) and head circumference (HC) at delivery based on the constitutional growth potential. SD scores were calculated according to a customized model that takes into account maternal variables (preconception weight, height and parity), paternal variables (height) and newborn variables (sex and gestational age at birth).22

Other covariates

At weeks 10-13 and 28-32 the pregnant women completed two detailed in-person questionnaires on anthropometric and socio-demographic characteristics and life style variables, and two semi-quantitative food frequency questionnaires (FFQs), as further described elsewhere.23

The following maternal variables were a priori considered to be included in the models: age (years), height (cm), pre-pregnancy body mass index (BMI, Kg/m2), zone of residence (rural and non-rural), country of origin (Spain and others), education (up to primary, secondary, and university studies), employment during pregnancy (yes and no), social class defined according to the most privileged occupation during pregnancy of the mother or the father using a widely employed Spanish adaptation of the international ISCO88 coding system (class I: managerial jobs, senior technical staff, and commercial managers; class II: skilled non-manual workers; and class III: manual workers), parity (none and ≥1 births), consumption of tobacco at the beginning of pregnancy (yes and no), passive smoking in at least two environments from among the home, workplace and restaurants/leisure areas (yes and no), season of last menstrual period, carpets at home (yes and no), curtains at home (yes and no), type of mattress (foam, box spring, latex, and other types), frequency of housekeeping (1 times/week), vacuum cleaner use (1 times/week), resident density, number of hours per week of television use, previous lactation (none, ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download