Mycotoxin health risk assessment modelling among maize-subsistence farmers living in Centane, Eastern Cape Province, South Africa

Abstract

Harmful mycotoxins such as fumonisin B (FB), deoxynivalenol (DON), and zearalenone (ZEA), produced by ubiquitous food-borne fungal species, are known to contaminate maize crops from subsistence farming areas in Centane, Eastern Cape Province (EC), South Africa. The daily consumption of home-grown maize in these areas has resulted in FB exposure of five to ten times above the recommended provisional maximum tolerable daily intake (PMTDI) of 2 µg kg-1 body weight day-1 , as set by The Joint FAO/WHO Expert Committee on Food Additives. From a public health perspective, not only can these mycotoxins cause various human diseases but also impact food security. For this purpose, mycotoxin risk assessment among maizesubsistence farmers is critical to ensure evidence-based risk management. Currently, in resources poor settings, a deterministic risk assessment approach remains the easiest and most accessible. This approach, based on epidemiological data, includes using a total mean mycotoxin level (µg kg-1 ) multiplied by the individual total mean dry/raw maize intake in g day-1 divided by body weight (kg) and expressed as a probable daily intake (PDI). The resultant PDI is thereafter compared to the relevant mycotoxin PMTDI to determine the risk. However, this type of assessment remains one-dimensional and can only quantify risk. To address this limitation, the current study was aimed at developing a mycotoxin risk assessment model that not only quantifies the risk of exposure to multi-mycotoxins (FB, DON, and ZEA) but it is also interactive. The new model was based on the Mycotoxin Risk Assessment Model (MYCORAM) that was originally developed for South African commercial maize consumers and referred to as the MYCORAM-II. The overall purpose of the MYCORAM-II is to assess the percentage of maize consumers above the relevant mycotoxins PMTDI. In the current study, a multiphase study design was used to develop and evaluate the model consisting of phase 1) dietary maize intake, body weight data collection, phase 2) the development of dietary dry maize intake per body weight categories, phases 3 and 4) mycotoxin levels data and the development of the MYCORAM-II, phase 5) evaluation of the MYCORAM-II using published mycotoxin levels in home-grown maize from Centane, and phase 6) to assess the appropriateness of the national maximum safety levels as set by the South African Department of Health for FB and DON in dry maize. The development of the model was limited to the availability of secondary data, its validity, and the type of data that has been collected and published from studies conducted over the past 10 years in rural maize-subsistence farming areas in EC, South Africa. Applying mycotoxin levels in home-grown maize from Centane to the MYCORAM-II indicated that for total FB between 80-90% of maize, consumers were above its PMTDI based on overall higher levels of FB. In comparison, DON and ZEA with their lower levels in home-grown maize resulted in a lower percentage of consumers above the respective PMTDIs. The MYCORAM-II also indicated that the South African national maximum safety levels will not protect subsistence maize consumers between (89% and 90% will be at risk). In addition, the MYCORAM-II indicated differential risk scenarios when using published data from African countries. Overall, the MYCORAM-II highlighted the distinct roles of FB levels in home-grown maize, the number of positive samples, and maize intake practices. Despite the limitation of the MYCORAM-II, its simplicity and interactive function render it useful to describe multi-mycotoxin exposure among subsistence maize farmers not only in Centane, EC, but also in Africa. Its application can also be expanded towards developing population-specific safety levels, facilitating decision-making during risk management, and the evaluation of appropriate public health interventions aimed at reducing mycotoxin exposure.