The Association of Arsenic Exposure and Metabolism Biomarkers with Subclinical Measures of Liver Disease: a cross-sectional investigation in the Multi-Ethnic Study of Atherosclerosis.

Long-term exposure to arsenic (As) can cause many health effects and As metabolism is key in the detoxification and elimination of As. We hypothesize that hepatic steatosis might affect As metabolism efficiency. We evaluated the association of As exposure and As metabolism biomarkers with hepatic steatosis. In this cross-sectional analysis, we analyzed data from 3,577 participants in the Multi-Ethnic Study of Atherosclerosis (MESA), an ethnically diverse US adult population, with urinary metals and liver CT scan data available. We measured total As and As species in urine, and summarized As methylation biomarkers as inorganic (iAs%), monomethylarsonic acid (MMA%), and dimethylarsinic acid (DMA%). The ratio of liver-to-spleen (L/S) Hounsfield units (HU) < 1.0 and liver attenuation < 40 HU were used to assess the presence and severity of liver fat content. We used logistic regression to estimate the odds ratio (OR) (95%CI) of steatosis per higher interquartile range (IQR) of log-transformed total urinary As levels (ΣAs, μg/L), and per 5% higher in log-transformed As species (iAs%, MMA%, DMA%). The adjusted ORs (95%CI) for L/S < 1.0 were 1.09 (0.91, 1.30) per higher IQR of ΣAs, and 0.80 (0.68, 0.95), 0.69 (0.61, 0.77), and 1.24 (1.15, 1.34) per 5% higher iAs%, MMA%, and DMA%, respectively. The corresponding ORs (95% CI) for HU < 40 were 0.99 (0.75, 1.30) per higher IQR of increase ΣAs, and 0.70 (0.50, 0.91), 0.65 (0.55, 0.78), and 1.31 (1.16, 1.48) per 5% higher iAs%, MMA%, and DMA%, respectively. The associations were consistent by self-reported race/ethnicity and sex. Arsenic methylation capacity, but not exposure, was associated with the prevalence of hepatic steatosis. Studies are needed to examine the longitudinal association between the progression of hepatic steatosis with As metabolism biomarkers and As-related disease. https://doi.org/10.1289/EHP16728.