Because of their lipophilic character, persistent organic pollutants are absorbed and transported to fatty tissues in fish and marine mammals. Uptake of POPs can occur through exposure from sediments in water or via consumption of smaller fish by predatory species (Geyer et al., 2000).
Farmed fish are exposed to these contaminants to the extent that they are present in feed (Hites et al., 2004a). Recently, Hites et al. (2004a) found that, perhaps because of their higher fat levels, some farmed salmon contain significantly higher concentrations of certain organochlorine contaminants, including PCBs, than wild-caught salmon. In addition, PCB concentrations in samples of commercial salmon feed purchased in Europe were higher than those in samples purchased in North and South America, suggesting that regional differences in the composition of feed contribute to regional differences in the PCB concentrations in farmed salmon. The mean wet weight concentration of PCBs in farmed salmon was 50 ng/g or below (Hites et al., 2004a), regardless of source, and thus below the Food and Drug Administration (FDA) action level of 2 ppm for PCBs in food. Using the US EPA risk assessment for PCB and cancer risk, Hites et al. (2004a) concluded that, given the PCB levels in the fish samples, a consumer’s risk will not be increased if consumption is limited to no more than 1 meal per month of farmed salmon. Given the substantial regional differences found in PCB levels, however, these analyses demonstrated the importance for the consumer of knowing whether a fish was farmed or wild-caught and also its region of origin.
In a subsequent paper, the same group of investigators reported a quantitative analysis of competing risks and benefits associated with consuming farmed Atlantic and wild-caught Pacific salmon, for both cancer and noncancer end points (Foran et al., 2005b). Sixteen organic contaminants were considered. A benefit/cancer risk ratio was calculated for cancer using cancer slope factors developed by the US EPA (assuming that a 1×10−5 risk is acceptable) and a benefit/noncancer risk ratio using reference doses established by the US EPA. Foran et al. (2005b) concluded that neither farmed nor wild-caught salmon can be consumed in quantities that would provide 1 g/day of EPA/DHA while still maintaining an acceptable level of carcinogenic risk (1×10−5). In contrast, they determined that based on the benefit/noncarcinogenic risk ratio, wild-caught salmon could be consumed in amounts consistent with EPA/DHA intake levels recommended by the American Heart Association (see Chapter 2).
As expected, however, the results differed for farmed and wild-caught salmon. Consuming farmed salmon in amounts that provides 1 g/day of EPA/DHA would produce a cumulative cancer risk that is 24 times the acceptable cancer risk level. For wild-caught salmon, the cumulative cancer risk would be eight times the acceptable level. Both farmed and wild-caught