Abstracts: ISEE 20th Annual Conference, Pasadena, California, October 12–16, 2008: Contributed Abstracts
Previous studies have investigated that the relationship between maternal mercury level and birth weight. The aim of the study was to investigate the effect of maternal mercury level on birth weight. Also, we examined the relationship among environmental factors, mercury level, and birth weight using a structural equation model.
The collaborating multi-centers, Seoul (metropolitan area), Ulsan (industrial area), and Cheonan (medium-sized urban area), for a prospective cohort study of Mothers and Children’s Health and Environment (MOCHE) have been built up in 2006, and we enrolled 770 women before their second trimester of their pregnancy along with their spouses in 2006–2007. A trained nurse interviewed participants to record general information on demographic and socioeconomic factors, medical history, complications of current gestation, health behaviors, and environmental factors. We collected blood and urine samples to measure biomarkers for environmental exposures including heavy metals. We used the atomic absorption spectrophotometer to measure mercury level in blood. Because of skewed data, we changed the maternal mercury level by log-transformation. A structural equation model described the sequential relationships among environmental factors, mercury level and birth weight. Model fitting was assessed by various indices, such as chi-square, x2/df, Root mean square error of approximation (RMSEA), Tucker-Lewis index(TLI), and normed fit index(NFI). In these models, observed variables are considered manifestations of a limited number of causally related latent variables, so this approach is useful for analyzing multidimensional epidemiologic data. We used AMOS 6.0 (SPSS Inc.) for the SEM analysis. Mercury levels of 631 pregnant women was analyzed. In addition, the pregnancy outcome records of 440 individuals were collected for this study.
The mean weight of babies was 3261 ± 429.54(g), and the mean concentration of mercury level in maternal blood during the first trimesters was 4.45 ± 3.78(μg/l). The path from fish consumption prior to pregnancy to maternal mercury level ran in a positive direction (0.27, P = 0.01). The path in the model showed that mercury level directly acted on the birth weight reduction. Also, after considering environmental factors, we examined that fish consumption might cause increases of mercury levels. Birth weight was indirectly decreased by fish consumption through mercury level.
This study suggests that fish consumption prior to pregnancy was significantly associated with maternal mercury level. Our results are consistent with numerous studies of environmental factors and birth weight. The finding of this study using a structural equation model (SEM) supports the construction of national policy for environmental health management.