Mineral oil can enter the food chain intentionally (e.g. additives) or unintentionally (e.g. residues from printing inks, recycling, machine oils,…). Over the last few years, contamination of food with mineral oil has been reported. However, occurrence data on mineral oil in food is rather scarce and the hazards of mineral oil are not sufficiently characterised. In this project, Sciensano assessed the potential health risks associated with exposure to mineral oil.
Mineral oil can enter the food chain via different routes. In order to obtain more insights in the extent of the mineral oil food contamination problem in Belgium, we conducted an extensive market survey including a wide variety of matrices. As saturated hydrocarbons (MOSH) and aromatic hydrocarbons (MOAH) have a different toxicological profile, we quantified the two fractions separately using an online combination of liquid chromatography and gas chromatography with flame ionisation detection (online LC-GC-FID). We then compared the detected MOSH and MOAH concentrations to the action thresholds set by the Scientific Committee (SciCom) of the Belgian Food Safety Agency (FASFC) in 2017. Only 1 sample exceeded the threshold for MOSH, while the threshold for MOAH was exceeded in 23 samples. For the samples exceeding the action threshold, further investigation is needed to identify the contamination source.
Next, we assessed the dietary exposure of the Belgian population to mineral oil by combining the analytical results with consumption data of the Belgian Food Consumption Survey. Overall, the exposure to MOSH and MOAH was much lower compared to the results previously reported. Afterwards, we assessed the possible risks for the Belgian population by combining the outcome of the exposure assessment with hazard-related information. Based on the margin of exposure (MOE) approach, we concluded that there is no health concern related to MOSH exposure for the Belgian population. For MOAH, which may be potentially genotoxic and carcinogenic, no adequate dose-response carcinogenicity data are available and consequently, we could not perform a MOE-based risk assessment. We applied 2 worst‐case scenarios to obtain some preliminary insights in the possible health risks associated with MOAH exposure. Based on these scenarios, we could not exclude adverse human health effects highlighting the need for more data to perform a more adequate risk assessment.
In order to fill some of the existing gaps in the toxicological knowledge on MOAH, we collected data on their endocrine activity and genotoxic potential. We obtained several interesting results, including further evidence that highly alkylated polycyclic aromatic hydrocarbons (PAHs) (which are important constituents of the MOAH fraction) are not metabolized into DNA‐reactive compounds due to steric hindrance. Furthermore, preliminary results suggest the possible use of bioassays based on activation of the Aryl receptor as a fast screening procedure to detect the presence of MOAH, but more research is needed to further investigate this hypothesis.