The Belgian PCB/dioxin incident: analysis of the food chain contamination and health risk evaluation.

The Belgian PCB incident occurred at the end of January 1999 when a mixture of polychlorinated biphenyls (PCBs) contaminated with dioxins was accidentally added to a stock of recycled fat used in the production of animal feeds. Although signs of poultry poisoning were noticed by February, 1999, the source and the extent of the contamination were discovered only in May 1999, when it appeared that more than 2500 farms could have been supplied with contaminated feeds. This resulted in a major food crisis, which rapidly extended to the whole country and could be resolved only by the implementation of a large PCB/dioxin food monitoring program. Screening for PCB contamination was based on the determination of the seven PCB markers. When PCB concentrations exceeded the tolerance levels of 0.1 (milk), 0.2 (poultry, bovine, and pig meat), or 1 (animal feed) microg/g fat, dioxins (17 PCDD/Fs congeners) were also determined. At the end of December 1999, the database contained the results of more than 55,000 PCB and 500 dioxin analyses. The study of PCB levels and profiles in contaminated feeds delivered to poultry or pig farms confirmed that the Belgian PCB incident was due to a single source of PCB oil introduced into the food chain at the end of January 1999. This PCB oil had a congeners pattern closely matched to a mixture of Aroclor 1260/1254 in the proportion 75/25. The total amount of PCBs added to recycled fats was estimated at 50 kg (sum of the seven markers) or approximately 150 kg total PCBs, which corresponds to about 100 liters of PCB oil. This PCB mixture contained about 1g TEQ dioxins (more than 90- contributed by PCDFs) and about 2g TEQ dioxin-like PCBs. The proportions of PCB 52 and 101 congeners were fairly constant in animal feeds, excluding the possibility of secondary contamination due to fat recycling from contaminated animals. The highest concentrations of PCBs and dioxins were found in poultry and especially in the reproduction animals (hens and chicks), which showed the classical manifestations of chick edema disease. The pigs were also affected but to a lesser extent and no sign of intoxication was observed. The study of PCB/dioxin patterns and of the PCB:dioxin ratios revealed major differences in the metabolism of these compounds by farm animals. Whereas the PCBs:dioxins ratio was fairly constant in all poultry products with a mean value similar to that found in contaminated feeds (50,000), in pigs this ratio was both much higher and more variable (values up to 10,000,000), reflecting a faster elimination of dioxins than PCBs in these animals. These metabolic differences also emerged from the PCB and dioxin patterns which were altered much more in pigs than in poultry. Although the most contaminated food products (chicken meat) had PCB and dioxin levels more than 100 times above maximal recommended values, it is unlikely that this incident could have caused adverse effects in the general population of Belgium. A doubling of the PCB and dioxin burden of the young adult population would require the consumption of, respectively, 10 and 20 highly contaminated meals. In view of the very limited proportion of the poultry chain effectively contaminated during the incident (around 2%), such an extreme scenario was quite improbable for the general population except perhaps for farmers consuming their own products. But even in that case, it would have meant going back to the levels in the 1980s or attaining the body burden of subjects regularly eating contaminated seafood.