| Sumario: | Aflatoxins and fumonisins commonly contaminate key food staples in tropical countries, causing recurring acute and chronic public health problems. The present study was conducted to assess the potential of a simple device designed for density-based sorting of maize for its potential to reduce aflatoxins and fumonisins in diverse samples of naturally contaminated maize. A cross sectional survey was conducted, analysing market maize samples (n=204) from eight counties in Western Kenya. A quarter (25%) of the maize samples were contaminated with aflatoxin B1 above the legal limit of 5 μg/kg and nearly half (48%) were contaminated with fumonisins at levels above the legal limit of 2 μg/g. Analysis of additional samples (n=24) from Meru County showed that contamination of maize with the two toxins was also common in Eastern Kenya. A simple density sorter was used to separate grain samples into heavy and light fractions. With an out-sort rate of 31%, density sorting was effective in separating maize by bulk density and 100-kernel weight (P<0.001). Bulk density was negatively correlated with aflatoxins in unsorted (r=-0.20, P<0.01) and heavy fractions (r=-0.32, P<0.01). Density sorting was effective at reducing fumonisins; for maize samples with >1 μg/g, the heavy (accepted) fraction had 66% lower fumonisins than the unsorted maize. After density sorting, the light and heavy fractions fumonisin levels differed by an average of 8.38 μg/g (P<0.001). However, sorting was not effective at significantly reducing aflatoxin levels in maize that was highly contaminated. A simple density sorting equipment that could be adopted by local small-scale millers has potential to reduce fumonisins in maize. Additional and complementary sorting technologies, such as size screening and spectral sorting might improve the effectiveness of reducing aflatoxins in maize.
|
|---|