RESUMO
A stochastic simulation model was developed to carry out the first quantitative risk exposure assessment of the mycotoxin level in cow's milk produced in Argentina. The prevalence and concentration of aflatoxin M1 (AFM1), deoxynivalenol (DON) and zearalenone (ZEA) were modeled at various stages through milk processes complying with Argentinean practices. Concentration of AFM1 (0.059ppb), DON (0.338ppb) and ZEA (0.125ppb) in dairy milk were estimated. The proportion of feed samples that exceeded the maximum level accepted by European regulations for AFB1, DON and ZEA were estimated at 25.07%, 0.0% and 8.9%, respectively. The percentage of milk samples that exceeded the maximum level accepted for AFB1 by the MERCOSUR (0.5ppb) and the European Union regulations (0.05ppb) were 0.81 and 32.65, respectively. The probability distribution of AFM1 concentration in milk was affected by the carry-over rate equations applied in the model. Mycotoxin levels in corn silage and concentrated feeds were the factors most correlated with mycotoxin concentrations in milk. Therefore, agricultural practices, crop management and feed production require prompt attention regarding mycotoxin issues.