The effect of dosing vehicle on arsenic bioaccessibility in smelter-contaminated soils

In vitro gastrointestinal (IVG) methods have been developed to provide an expedient and inexpensive means to estimate bioavailability of arsenic and other contaminants from ingestion of contaminated soil. Both in vivo and in vitro techniques have used a fasting model when determining Pb bioavailability/bioaccessibility as a conservative estimate of risk. Some IVG procedures have incorporated a dosing vehicle (DV) or food (i.e., milk) to simulate in vivo conditions. Potential differences in the bioaccessibility of contaminants between fasting and fed states remain a concern for those interested in adopting in vitro procedures for regulatory purposes. In this study, the effect of eliminating a dough-like DV on As bioaccessibility (BA), and this effect on the relationship between in vitro bioaccessible and in vivo relative bioavailability (RBA) As is determined. Also, the effect of phosphate from the DV on IVG BA is investigated. Two types of smelter-contaminated soils, calcine and iron slag, were used to examine the effect of dosing vehicle (DV) on BA determined by IVG. Dosing vehicle did not affect BA in the gastric extraction (GE) or intestinal extraction (IE) for 3 of the 5 calcinated contaminated soils. Inclusion of DV in the GE slightly increased BA for 2 of the 5 slag-contaminated soils. Increases in BA from DV may be attributed to ligand exchange of arsenate with phosphate. Strong relationships between BA and in vivo RBA As were found with or without DV. Bioaccessible As measured by the GE was strongly correlated with in vivo RBA As ( IVG without DV: r = 0.92, P < 0.01; IVG with DV: r = 0.96; P < 0.01). Similarly, BA measured by the IE was strongly correlated with in vivo RBA As ( IVG without DV: r = 0.90, P < 0.01; IVG with DV: r = 0.96, P < 0.01). The IVG method, with or without DV, is a reliable method to use as a rapid screening tool to provide an estimate of BA in contaminated soils. Further studies should be conducted to determine the influence of foodstuffs on BA for different types of As contaminated soil (i.e., non-smelter soil).