Development And Validation Of A Biomarker To Estimate Inorganic Arsenic Exposure In Populations With High Seafood Intake: Evidence From The Multi-Ethnic Study Of Atherosclerosis (MESA) And The National Health And Nutrition Examination Survey (NHANES)

Abstract

Introduction: In populations with low seafood intake, the sum of inorganic (iAs) and methylated (methylarsonate [MMA] and dimethylarsinate [DMA]) arsenic species (ƩAs) is used as the biomarker of iAs exposure. However, with moderate to high seafood intake, iAs exposure assessment is difficult as some seafood arsenicals contribute to DMA and maybe to MMA. We developed and validated a biomarker to estimate iAs exposure in populations with common seafood intake. Methods: The model was developed for ƩAs and individual species in 310 randomly selected MESA participants in 6 US cities and validated for DMA in 2,010 NHANES 2003-2006 participants (representative sample of US adults). We imputed urinary concentrations of iAs, DMA, MMA and ƩAs by regressing the original concentrations by arsenobetaine concentrations and extracting model residuals. To confirm that imputed biomarkers no longer reflected seafood intake but reflected exposure to iAs exposure, we estimated ratios of the geometric means of urinary arsenic with seafood and rice intake, respectively. Results: In MESA, median (IQR) for ƩAs was 8.1 (3.8, 15.5) and 7.5 (4.2, 13.2) µg/L before and after imputation, respectively. Compared to participants reporting rare/never seafood intake, those reporting more frequent seafood intake had higher concentrations for all the original urinary arsenic biomarkers. Using the imputed biomarkers, self-reported seafood intake or estimated n-3 fatty acids were no longer associated with urinary arsenic measures in MESA; self-reported intake of rice remained associated with our imputed arsenic biomarkers. In NHANES, the association between urinary DMA concentrations with self-reported seafood intake and estimated n-3 fatty acids practically disappeared with imputation. Conclusions: Correction for seafood exposure using the biomarker arsenobetaine can facilitate research about low-moderate levels of iAs exposure through water and food in general populations with frequent seafood intake.