Estimation of furan contamination across the Belgian food chain

Abstract

This paper provides an estimate of the furan content of Belgian foods. The objective of the study was to achieve the best food chain coverage with a restricted number of samples (n = 496). The geographic distribution, different market chains and labels, and consumption frequencies were taken into account in the construction of the sampling plan. Weighting factors such as contamination levels, consumption frequency and the diversity of food items were applied to set up the model. The very low detection capabilities (CC_β) of the analytical methods used (sub-ppb) allowed reporting of 78.2% of the overall dataset above CC_β and, in particular, 96.7% for the baby food category. The highest furan levels were found in powdered roasted bean coffee (1912 µg kg⁻¹) with a mean of 756 µg kg⁻¹ for this category. Prepared meat, pasta and rice, breakfast cereals, soups, and baby food also showed high mean furan contents ranging from 16 to 43 µg kg⁻¹. Comparisons with contamination surveys carried out in other countries pointed out differences for the same food group and therefore contamination levels are related to the geographical origin of food items.

Keywords:

• baby food
• coffee
• meat
• processed foods
• process contaminants
• exposure assessment

Acknowledgements

This study was funded by the Belgian Federal Public Service of Health, Food Chain Safety and Environment (Contract Number RT 06/01 FURA). The authors also want to acknowledge Christine Vinkx, Ria Nouwen and Isabel DeBoosere from the Federal Public Service of Health, Food Chain Safety and Environment; Katleen Baert from the Belgian Agency for Food Safety (AFSCA); Norbert Dekimpe, Bruno De Meulenear and An Adams from the University of Gent; and Guy Maghuin-Rogister from the University of Liege for their participation in the committee of experts.