Oh dear! Another report, (Ronald A. Hites, R.A. et al., (2004), Global assessment of organic contaminants in farmed salmon, Science, 303, 226-229), finding that farmed salmon contain Highly Chlorinated Hydrocarbons (HCHs) in the flesh. This is becoming an annual event. The topic was raised on the Listserve last March following the EWG report (Easton, M.D.L., Luszniak, D. & Von der Geest, D., (2002), Preliminary examination of contaminant loadings in farmed salmon, wild salmon and commercial salmon feed. Chemosphere, 46,1053_1074). In early 2001 was the report by Jacobs et al (Investigation of PCDDs, PCDFs and selected coplanar PCBs in Scottish farmed Atlantic salmon (Salmo salar), Organohalogen Compounds, 47, 338_341). What does the new study tell us that we don't already know? In fact very little, but in more detail. Hites and co-workers measured HCHs in 700 or so samples of farmed Atlantic salmon and wild Pacific salmon of several species. The authors reported they did not sample wild Atlantic salmon because "... few are available commercially..." , nor
farmed Pacific salmon " ... because they are not raised in any substantial amounts ...". I note that the Jacobs et al study included wild Atlantic salmon and the Easton et al study included farmed pacific salmon. The authors therefore compared farmed Atlantic salmon with wild Pacific salmon. Readers should decide for themselves whether this is an helpful or meaningful comparison.
The results showed that for all HCHs measured mean concentrations in the Atlantic salmon were significantly higher than in the Pacific salmon. Comparisons of data on HCH concentrations in foods is a bit of a minefield because of different ways of expressing concentrations. The Hites paper shows summary data in the form of figures and expressed on a wet weight basis except for what they list as 'dioxins' for which concentrations are expressed at TEQs wet weight basis. PCBs and dioxins exist as many isomers only a few of which are considered toxic and raw concentrations are weighted by Toxic Equivalency factors (TEQ) for the individual isomers to give an index expressing the relative toxicity of the totality of contaminants in the sample. Some PCB isomers are considered to have similar toxic effects as dioxins so the weighted PCB and dioxin concentration are usually added to give an overall toxicity value. Hites and co-workers have done this and their 'dioxin' values are combined TEQs for PCB and dioxin. Hites et al find that Atlantic salmon sampled in Europe and Scandinavia have 'dioxin' concentrations as TEQ wet weight basis of between 2 and 3 picog/g. Salmon purchased in the USA and Canada, which are assumed to be predominately of Chilean origin, are in the range 1-2 picog/g, and Pacific salmon are less than 0.5 picog/g. The values for Atlantic salmon are comparable with those found in other surveys of farmed salmon, though sometimes comparisons are not direct because some surveys express concentrations on a lipid basis rather than a wet weight basis and need to be corrected by the known or assumed lipid content of the sample. Hence my conclusion expressed above that the Hites et al paper has not told us anything we did not already know.
I consider it a major weakness of the Hites et al paper that the authors do not make any comparisons of their data with those from other surveys. In their bibliography they cite the Easton and the Jacobs studies I have referred to above, but do not summarise the results and do not cite any other reports. They do not report the results from food surveys in the UK that show that concentrations of PCBs/dioxins in foods are decreasing quite markedly over time; the estimated average dietary intake by adults in the UK has fallen from 1.8 picog/kg body wt/day in 1997 to 0.9 picog/kg body wieght/day in 2001. The authors should point out that concentrations of PCBs/dioxins in the environment, including the aquatic environment, are decreasing over time and risks to human health will reduce over time. The authors have not summarised and compared data on PCBs/dioxins in species other than salmonids some of which show concentrations higher that those reported for farmed Atlantic salmon.
Hites et al acknowledge that "Individual contaminant concentrations in farmed and wild salmon do not exceed U.S. Food and Drug Administration (FDA) action or tolerance levels for PCBs and dieldrin", but do not comment on the other HCHs. I have not attempted to find tolerable levels - if they exist - for the other HCHs, but as the authors do not comment on risks for these other contaminants I assume they are within limits. Though the authors cite US FDA tolerance levels, the concentrations of combined PCBs and dioxins expressed as TEQs are within the WHO recommendations which have been adopted in many countries including the UK. Hites et al have determined the risks from the US FDA model for fish advisories and have come up with the advice shown in Figure 4 in their report - less than one meal/month of farmed Atlantic salmon. They give no estimates of the confidence of these estimates. Smith, G.C., et al. (2002, Intake estimation of polychlorinated dibenzo_p_dioxins, dibenzofurans (PCDD/Fs) and polychlorinated biphenyls (PCBs) in salmon: the inclusion of uncertainty. Food Additives and Contaminants, 19, 770_778) have pointed to many uncertainties in estimating the health risks of consumption of PCBs/dioxins.
Am I getting cynical and suspicious in my old age, or do I detect a campaign behind this paper. The news about this study broke in the TV teletext a little before 10.00 p.m local time (GMT). I checked the BBC web page and found the news item which was clearly a press release. I was a bit suspicious of this as I would not expect the BBC news editors to be pick up a topic from Science on the day it was published. Though the item did not acknowledge the source it included a link to Hites at the University of Indiana and I asked for a copy of the copy. I was amazed to get it almost by return; the actual times on the e-mail messages are 10.04 GMT for the request and 10.08 GMT for the reply and attachment. The office in the University must have been expecting responses to the press release. I note that the authors of the paper acknowledge that "This research was initiated and supported by the Environmental Division of the Pew Charitable Trusts.". Note the 'initiated'. Environmental pressure groups have mounting a campaign against salmon farming for some years now and it seems to me this study was initiated as part of such a campaign rather that to objectively examine the risk to human health of contaminants in fish.
Peter Howgate
This archive was generated by hypermail 2b29 : Fri Jan 09 2004 - 13:16:32 PST