Risk Assessment for Essential Nutrients: An Example with Manganese. H. J. Clewell and M. E. Andersen, ICF Kaiser International, 602 E. Georgia Ave., Ruston, LA 71270
With essential elements such as manganese (Mn) that are toxic at higher doses, various regulatory bodies are required to set both recommended daily allowances (RDAs) or estimated safe and adequate daily dietary intakes (ESADDIs) for nutrition, on the one hand, and reference doses (RfDs) or reference concentrations (RfCs) for toxicity, on the other. Current risk assessment approaches estimate no observed adverse effect levels (NOAELs) from animal experiments or from epidemiological studies and then apply a variety of uncertainty factors. With essential elements, the application of these uncertainty factors can quickly provide exposure limits that equate to daily intake levels well below those required for adequate nutrition. Current risk assessment methods are also incapable of quantitatively considering the impact of essentiality in the case of systemic toxicity from inhalation of an essential element such as Mn. Novel risk assessment paradigms are required to establish consistent exposure guidelines for essential elements independent of the route of exposure. A potential strategy for devising a consistent standard-setting approach for potentially toxic essential nutrients is to restrict exposures by any route to levels that cause a minimal increase in blood or target tissue concentrations compared to the variation of blood or target tissue concentrations in normal individuals. We propose an approach that would regulate exposures by any route of administration to daily doses that cause an increase in average target tissue concentration of less than 10% of the variation of normal values. An example of such an approach is presented for the case of neurotoxicity from inhaled or ingested Mn. A PBPK model of Mn was developed to predict concentrations of Mn in the brain for both inhalation and drinking-water exposures to Mn, as a function of the amount of Mn present in the diet. This model was then exercised to estimate acceptable multimedia exposures to Mn, as well as to aid in the design of critical experiments to improve the accuracy of the estimates. We believe the strategy described here is appropriate not only for Mn but also for other nutrients, such as vitamin A, that are toxic in overdose conditions.