Euro wines carry dangerous amounts of heavy metals
October 31, 2008, by Karl Storchmann (Journal of Wine Economics)
I just got an email from a fellow wine economist from Kiel, Germany, that reads as follows. "We once had a farmer association president who claimed to welcome all food imports [to the European Union], as long as they carried a skull and cross-bones label. This is a skull and cross-bones label for European wines." He referred to an article entitled Heavy metal ions in wines: meta-analysis of target hazard quotients reveal health risks by Declan P. Naughton and Andrea Petroczi. The article was published in the "Chemistry Central Journal" today.
Naughton and Petroczi found that most European wine nations are exporting red and white wines with potentially dangerous levels of at least seven heavy metals. The authors refer to so-called "target hazard quotients" (THQ) which were developed by the U.S. Environmental Protection Agency (EPA) for the estimation of potential health risks associated with long term exposure to chemical pollutants. The THQ is a ratio between the measured concentration and the oral reference dose, weighted by the length and frequency of exposure, amount ingested and body weight. The THQ value is a dimensionless index of risk associated with long term exposure to chemicals based upon reference upper safe limits. A value of THQ<1 means that there is no health risk associated with intake or exposure. A limited number of THQ investigations have been reported in foodstuffs with the focus being on estimating health risks associated with exposure to heavy metals found in seafoods, and in one case breast milk. Calculations of THQ values for seafoods were deemed high as many species accumulate heavy metals and other pollutants in their tissues. Many of the reported THQ values calculated from metal contaminants in seafood range from a safe level (<1) to a level of concern (typically THQ >1 to <5) with a small number being above 10.
If you think a THQ value of 5 in seafood is a health risk check the numbers for wine. Naughton and Petroczi found that typical wines have a THQ ranging from 50 to 200 per glass with some up 300.
Of course there is a wide range in THQ values from wine to wine and from country to country. Judging by the lowest THQ levels found this is the order of countries from worst to best: Portugal > Austria > France > Spain > Czech Republic >Hungary > Germany > Serbia.
Judging by the highest THQ levels found this is the order of countries from worst to best: Hungary >Slovakia > France >Austria > Spain > Germany > Portugal > Greece > Czech Republic > Jordan >Macedonia > Serbia.
In the Figure below Naughton and Petroczi report their findings by country of origin and separately for men and women (since their THQs differ due to differences in body weight and life expectancy). [Note that the authors say nothing about the specific wines in their sample nor do we know whether the sample is representative].
There is not a lot we know about the impact of heavy metals on health (there is a suspected association of manganese with Parkinson's disease). Whatever the critical THQ is, it is worrisome that (1) nobody had reported THQs for wine yet and (2) that we do not know where the contamination comes from. Naughton and Petroczi consider the influence of grape variety, soil type, geographical region, insecticides, containment vessels as possible source.
Here is the link to their Full Paper. In contrast to many economics papers, natural science papers are short!
I just got an email from a fellow wine economist from Kiel, Germany, that reads as follows. "We once had a farmer association president who claimed to welcome all food imports [to the European Union], as long as they carried a skull and cross-bones label. This is a skull and cross-bones label for European wines." He referred to an article entitled Heavy metal ions in wines: meta-analysis of target hazard quotients reveal health risks by Declan P. Naughton and Andrea Petroczi. The article was published in the "Chemistry Central Journal" today.Naughton and Petroczi found that most European wine nations are exporting red and white wines with potentially dangerous levels of at least seven heavy metals. The authors refer to so-called "target hazard quotients" (THQ) which were developed by the U.S. Environmental Protection Agency (EPA) for the estimation of potential health risks associated with long term exposure to chemical pollutants. The THQ is a ratio between the measured concentration and the oral reference dose, weighted by the length and frequency of exposure, amount ingested and body weight. The THQ value is a dimensionless index of risk associated with long term exposure to chemicals based upon reference upper safe limits. A value of THQ<1 means that there is no health risk associated with intake or exposure. A limited number of THQ investigations have been reported in foodstuffs with the focus being on estimating health risks associated with exposure to heavy metals found in seafoods, and in one case breast milk. Calculations of THQ values for seafoods were deemed high as many species accumulate heavy metals and other pollutants in their tissues. Many of the reported THQ values calculated from metal contaminants in seafood range from a safe level (<1) to a level of concern (typically THQ >1 to <5) with a small number being above 10.
If you think a THQ value of 5 in seafood is a health risk check the numbers for wine. Naughton and Petroczi found that typical wines have a THQ ranging from 50 to 200 per glass with some up 300.
Of course there is a wide range in THQ values from wine to wine and from country to country. Judging by the lowest THQ levels found this is the order of countries from worst to best: Portugal > Austria > France > Spain > Czech Republic >Hungary > Germany > Serbia.
Judging by the highest THQ levels found this is the order of countries from worst to best: Hungary >Slovakia > France >Austria > Spain > Germany > Portugal > Greece > Czech Republic > Jordan >Macedonia > Serbia.
In the Figure below Naughton and Petroczi report their findings by country of origin and separately for men and women (since their THQs differ due to differences in body weight and life expectancy). [Note that the authors say nothing about the specific wines in their sample nor do we know whether the sample is representative].
There is not a lot we know about the impact of heavy metals on health (there is a suspected association of manganese with Parkinson's disease). Whatever the critical THQ is, it is worrisome that (1) nobody had reported THQs for wine yet and (2) that we do not know where the contamination comes from. Naughton and Petroczi consider the influence of grape variety, soil type, geographical region, insecticides, containment vessels as possible source.
Here is the link to their Full Paper. In contrast to many economics papers, natural science papers are short!
I'm really surprised Italy isn't found on the worst list. In the Napoli area the mafia has a death grip on waste disposal and have happily buried or burned garbage wherever they like. The result is high levels of toxins that have seen local cancer rates well above Italy's average. Of course, household waste probably doesn't have a lot of heavy metals, but all that burning plastic should imbibe the wine with a unique flavor of dioxins.
Reply to this
This paper has too many errors in it to count. I will list just a few:
1. The Total Hazard Quotient (THQ) is an index created for a specific and very different material soils at hazardous waste sites. It is wrong to use it for food or wine.
2. The purpose of the THQ is to determine whether under worst-case conditions risks are so low that a formal risk assessment is NOT necessary. It is NOT a measure of risk, nor do high THQ values imply the existence of risk.
3. One cannot rank wines by THQ -- or soils at hazardous waste sites -- and correctly rank-order their risks.
4. None of the components of the THQ is a measure of risk. Each component is a highly precautionary (i.e., extreme) screening level (SL). When multiple precautionary SLs are summed, the result is yet more precautionary (i.e., more extreme). This has some value for deciding whether a formal risk assessment is NOT warranted, but it has no value for estimating or comparng risks.
4. Each SL has uncertainties of 10x to 10,000x built-in. Therefore, it is analytically silly and seriously misleading to report them with as many as five significant figures.
Even the authors caution against interpreting their results as you have done. "It is notable that the measured levels of metals do not necessarily reflect the risk to health..." Unfortunately, they follow this correct statement with a nonscientific one that suggests a policy agenda ("Therefore the THQ values are better measures of the levels of concern").
This paper should have been rejected by the journal. This likely occurred because the editor did not understand the content and did not know that the authors had misapplied a tool of rough-and-ready regulatory toxicology. This is a useful reminder to all journal editors that they should venture into unfamiliar areas with great caution.
The journal itself is very new -- volume 2. It is a very young, online publication trying to break into a crowded market. It is too young to have established a track record. It has ~40 (!) section editors and a huge editorial board, the role of which is not entirely clear. These features ensure that quality will be highly variable under the best of conditions. Finally, note that the journal is funded by payments received by authors (£800, €1010, or US$1,270) per accepted article, a known source of conflict of interest in scholarly publication.
Reply to this
The penultimate paper is missing a word that reversed its meaning. It should have read:
"This paper should have been rejected by the journal. This likely DID NOT OCCUR because the editor did not understand the content and did not know that the authors had misapplied a tool of rough-and-ready regulatory toxicology. This is a useful reminder to all journal editors that they should venture into unfamiliar areas with great caution."
Reply to this
Richard, it seems as though there are more people that deem this paper exaggerating or even flawed. I am not a chemist and cannot judge this. But check this article on today's Wine Spectator online site: http://www.winespectator.com/Wine/Features/0,1197,4702,00.html
Reply to this
The suggestion in the WS article that THQ is used for seafood is facially wrong. The index was created by EPA, and EPA has no jurisdiction over seafood.
The implicit reminder that "the dose makes the poison" is something I did not mention in my initial comment, but it surely is not hard for wine economists to appreciate that standard precautionary assumptions for drinking water consumption (2 liters per day) might not be appropriate here. At least, at that consumption level, heavy metals are the least of a person's health risks.
The statement that U.S. Alcohol and Tobacco Tax and Trade Bureau (TT
That the reputations of the authors is formidably damaged goes without saying, but I don't know if they had reputations to protect. Nonetheless, this shows how a new journal can destroy its own future reputation just to get a few headlines.
Reply to this