Development of analytical methods to quantify furan and alkyl furans in Belgian foodstuffs: from contamination levels, formation processes to preliminary risk assessment of the Belgian population

The issue of furan in food was raised in 1995, after IARC classified it as possibly carcinogenic to humans. This issue intensified after the publication of the EFSA report in 2017, which indicated the co-occurrence of furan with several furan derivatives with similar potential toxicity. To date, the available data on their contamination levels, the risk associated with their presence in food, and several other aspects are rather limited to the furan compound alone.
This work was carried out to extend the knowledge to alkyl furans and had five main objectives:
(1) to develop analytical methods for the determination of furan and its derivatives (2-methylfuran, 3-methylfuran, 2-ethylfuran, 2,5-dimethylfuran and 2,3-dimethylfuran) in food.
(2) Due to the high volatility of these compounds, especially furan, they are susceptible to be released into the ambient air, which is an additional source of exposure. Therefore, the second objective was to develop an analytical method to measure airborne furans.
(3) To determine the levels of these compounds in Belgian food.
(4) To study the mechanism of formation in coffee, the food most contaminated with furans, and to evaluate the changes in their concentrations during the different stages of coffee and chocolate preparation.
(5) To prioritise food matrices according to hazard ratio and hazard index for future risk assessment, and to prepare a survey on coffee consumption habits.
Achievements: In this thesis (1) we have optimised and validated the HS-SPME GC-MS analytical method for the measurement of furans in food. The results showed sensitivity and accuracy to meet EU requirements. (2) We have developed a TD-GC-MS method for the measurement of furan in air. The results of the measurement, carried out in an experimental chamber, show that the levels of furan emitted into the air during coffee brewing are low and that further measurements are required to consider chronic exposure. (3) A large sampling plan of food samples was designed to mimic Belgian consumption as closely as possible. The results of the analysis of 1003 samples show that coffee is the most contaminated food with furan compounds, both in terms of contamination levels and the frequency of presence of furans. The next food groups are canned fish, canned meat and baked products due to the abundance of 2-ethylfuran and 2-methylfuran respectively. (4) We have studied the variation of furan concentrations at different stages of coffee and chocolate processing. Regarding coffee, the study of furans formation during coffee roasting illustrates the important role of sucrose in the formation of these compounds. Acrylamide was also included in this study and shows a negative correlation with furan in terms of concentration. The study of furans formation during coffee brewing suggests that there is no evidence of furan formation. With regard to chocolate processing, both tempering and laser printing can increase the total furan concentration by up to 24% and 31% respectively. (5) We conducted an online survey on coffee brand loyalty among Belgian coffee consumers. The consumption dataset generated from this survey (1930 participants) shows that taste and price are the main factors influencing the choice of a particular coffee brand. The survey also illustrates the influence of socio-demographic characteristics on coffee consumption and shows that the median coffee consumption is 3 cups per day. Finally, we used the hazard ratio concept in order to prioritise food groups for future risk assessment. The results show a higher potential hazard for coffee compared to other food groups.