Abstract

Abstract Objectives To identify foods associated with blood metal concentrations during the 1st trimester of pregnancy. Methods In early pregnancy (11.3 ± 2.8 gestational week), 1206 women from Project Viva completed food frequency questionnaire and provided a blood sample to measure essential and non-essential metals. This analysis included 135 food items and 14 metals (As, Cd, Hg, Pb, Co, Cu, Mg, Mn, Se, Zn, Ba, Cs, Sb, Sn) in red blood cells (RBC). We perform a diet-wide association study (DWAS) of all bivariate relationships between each food-metal combination. Additionally, we used machine learning models with elastic-net regularization to estimate the % variation in metal concentrations explained by diet. Results Mean (standard deviation [SD]) age was 32.5 (4.5) years and pre-pregnancy BMI was 24.8 (5.4) kg/m2. Most participants were white (76.0%), not current smokers (88.5%), completed a college degree (75.5%), and had a household income >$70 K/year (66.8%). Compared to other US-based cohorts, the overall diet quality of participants was higher, and levels of toxic metals were lower. DWAS identified significant associations between several food items with As, Hg, Pb, Cs and Se; among the positives associations, for example, each SD increase in fresh fruit, white rice, and seafood was associated with 4%, 8%, and 18% higher median concentrations of As, respectively; and fresh fruits (12%), eggs (12%), white rice (15%), green vegetables (17%), and seafood (64%) was associated with Hg. Elastic net models identified significant positive and negative dietary predictors of As, Cs, and Hg, but not other metals. The % variation explained by the selected dietary variables was 21% for As, 7% for Cs, and 11% for Hg. Conclusions DWAS provides systematic evaluation of diet-metal relationship and can uncover potential dietary sources of prenatal metal exposure. In this population, we found positive associations between fruit, green vegetable, white rice and seafood consumption with As and Hg. Prenatal diet may be an important source of metals exposures, including As, Hg, and Cs. Funding Sources This work was supported by the US National Institutes of Health grants R01HD034568, UH3OD023286, U2CES026561and U2CES026555, and metal analysis was conducted by the Children's Health Exposure Analysis Resource (CHEAR award #2017–1740).

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call