A quantitative risk assessment for fumonisins B1 and B2 in US corn

Quantitative risk analysis permits modifying risk estimates with changes in variables such as exposure. This analysis for exposure to the mycotoxin fumonisin describes the magnitude of adverse effects, variability in the population and uncertainty of models as a range of possible outcomes. The most sensitive adverse response in rats, nephrotoxic lesions, was used for the dose-response analysis. Dietary intake of corn products was estimated from a 3-day consumption survey. Levels of corn in each product were estimated by standard methods. Fumonisin levels in corn products were estimated from Food and Drug Administration (FDA) surveillance data and distributions of fumonisin consumption were modelled for each eater in the survey population. Uncertainty for predictions made from each model and uncertainty resulting from model selection were described. Results of the dose-response and exposure analyses were assimilated in a two-dimensional Monte-Carlo simulation. Distributions representing variability and uncertainty were iteratively selected to form an array of estimates of the risk. On the basis of this analysis, current dietary levels of fumonisin would not result in renal lesions even at upper levels of exposure. To avoid toxicity at much higher doses, limiting corn intake would be more effective than would limiting the level of fumonisin in corn.