摘要:Background Deoxynivalenol (DON) is a toxic fungal metabolite that frequently contaminates cereal crops. DON is toxic to animals, but the effects on humans are poorly understood, in part because exposure estimates are of limited precision. Objectives In this study we used the U.K. adult National Diet and Nutrition Survey to compare 24-hr urinary DON excretion with cereal intake. Methods One hundred subjects were identified for each of the following cereal consumption groups: low (mean, 107 g cereal/day; range, 88–125), medium (mean, 179 g/day; range, 162–195) and high (mean, 300 g/day; range, 276–325). DON was analyzed in 24-hr urine samples by liquid chromatography–mass spectrometry after purification on immunoaffinity columns. Results DON was detected in 296 of 300 (98.7%) urine samples. Cereal intake was significantly associated with urinary DON ( p < 0.0005), with the geometric mean urinary levels being 6.55 μg DON/day [95% confidence interval (CI), 5.71–7.53]; 9.63 μg/day (95% CI, 8.39–11.05); and 13.24 μg/day (95% CI, 11.54–15.19) for low-, medium-, and high-intake groups, respectively. In multivariable analysis, wholemeal bread ( p < 0.0005), white bread ( p < 0.0005), “other” bread ( p < 0.0005), buns/cakes ( p = 0.003), high-fiber breakfast cereal ( p = 0.016), and pasta ( p = 0.017) were significantly associated with urinary DON. Wholemeal bread was associated with the greatest percent increase in urinary DON per unit of consumption, but white bread contributed approximately twice as much as wholemeal bread to the urinary DON levels because it was consumed in higher amounts. Conclusion The majority of adults in the United Kingdom appear to be exposed to DON, and on the basis of the urinary levels, we estimate that some individuals may exceed the European Union (EU) recommended maximum tolerable daily intake of 1,000 ng DON/kg (bw). This exposure biomarker will be a valuable tool for biomonitoring as part of surveillance strategies and in etiologic studies of DON and human disease risk.