Consumption of freshwater fish: A variable but significant risk factor for PFOS exposure

PFOS, PFOA, PFNA and PFHxS are the PFAS substances that currently contribute most to human exposure, and in 2020 the European Food Safety Authority (EFSA) presented a draft opinion on a tolerable intake of 8 ng/kg/week for the sum of these four substances (equaling 0.42 mu g/kg if expressed as an annual dose). Diet is usually the dominating exposure pathway, and in particular the intake of PFOS has been shown to be strongly related to the consumption of fish and seafood. Those who eat freshwater fish may be especially at risk since freshwater and its biota typically display higher PFOS concentrations than marine systems. In this study, we estimated the range in PFOS intake among average Swedish "normal" and "high" consumers of freshwater fish. By average we mean persons of average weight who eat average-sized portions. The "normal consumers" were assumed to eat freshwater fish 3 times per year, and the "high consumers" once a week. Under these assumptions, the yearly tolerable intake for "normal" and "high" consumers is reached when the PFOS concentrations in fish equals 59 and 3.4 mu g per kg fish meat. For this study, PFOS concentrations in the muscle tissue of edible-sized perch, pike and pikeperch were retrieved from three different Swedish datasets, covering both rural and urban regions and a total of 78 different inland waters. Mean PFOS concentrations in fish from these sites varied from 0.3 to 750 mu g/kg. From the available data, the annual min-max dietary PFOS intake for male "normal consumers" was found to be in the range 0.0021-5.4 mu g/kg/yr for the evaluated scenarios, with median values of 0.02-0.16 mu g/kg/yr. For male "high consumers", the total intake range was estimated to be 0.04-93 mu g/kg/yr, with median values being 0.27-1.6 mu g/kg/yr. For women, the exposure estimates were slightly lower, about 79% of the exposure in men. Despite highly variable PFOS concentrations in fish from different sites, we conclude that the three most commonly consumed freshwater species in Sweden constitute an important source for the total annual intake even for people who eat this kind of fish only a few times per year. The analyses of PFOA, PFNA and PFHxS showed values which were all below detection limit, and their contribution to the total PFAS intake via freshwater fish consumption is negligible in comparison to PFOS