Urinary Arsenic in Human Samples from Areas Characterized by Natural or Anthropogenic Pollution in Italy

Abstract

Arsenic is ubiquitous and has a potentially adverse impact on human health. We compared the distribution of concentrations of urinary inorganic arsenic plus methylated forms (uc(iAs+MMA+DMA)) in four Italian areas with other international studies, and we assessed the relationship between uc(iAs+MMA+DMA) and various exposure factors. We conducted a human biomonitoring study on 271 subjects (132 men) aged 20–44, randomly sampled and stratified by area, gender, and age. Data on environmental and occupational exposure and dietary habits were collected through a questionnaire. Arsenic was speciated using chromatographic separation and inductively coupled mass spectrometry. Associations between uc(iAs+MMA+DMA) and exposure factors were evaluated using the geometric mean ratio (GMR) with a 90% confidence interval by stepwise multiple regression analysis. The 95th percentile value of uc(iAs+MMA+DMA) for the whole sample (86.28 µg/L) was higher than other national studies worldwide. A statistical significant correlation was found between uc(iAs+MMA+DMA) and occupational exposure (GMR: 2.68 [1.79–4.00]), GSTT gene (GMR: 0.68 [0.52–0.80]), consumption of tap water (GMR: 1.35 [1.02–1.77]), seafood (GMR: 1.44 [1.11–1.88]), whole milk (GMR: 1.34 [1.04–1.73]), and fruit/vegetables (GMR: 1.37 [1.03–1.82]). This study demonstrated the utility of uc(iAs+MMA+DMA) as a biomarker to assess environmental exposure. In a public health context, this information could be used to support remedial action, to prevent individuals from being further exposed to environmental arsenic sources.

Factors Affecting Arsenic Methylation in Contaminated Italian Areas

“…Variables considered in the literature associated with As methylation were selected, namely gender, smoking habits, alcohol consumption and BMI [ 23 ]. We also included some variables that emerged as informative in a recent study performed using SEpiAs data [ 70 ], i.e., the area of residence, water consumption, exposure to chemical and physical agents, dietary habits and some gene polymorphisms ( Table 2 ). Particularly, the variables for which more than two exposure classes were defined in the questionnaire were dichotomized as described in the “Value” column in Table 2 (exposure: yes/no, consumption: yes/no).…”