Paracelsus' poison

Paracelsus' poison

Are toxic Dioxin levels lurking in our weed killers? Follow up on the Four Corners report

Dioxin. That word conjures up images of Seveso, Yusho and Agent Orange. Catastrophic exposure to dioxin has resulted in a range of long term health issues and death from cancer. Dioxin is not a desired chemical, but is a side product of the manufacture of several chemicals, as well as being produced by natural processes such burning wood.

Dioxin is a contaminant of the herbicide 2,4,5-T and to a much lesser extent 2,4-D. Unlike exposure from major industrial accidents, conclusively showing that manufacture and use of herbicides is associated with long term health effects has been much harder.

For Australians, this issue became moot in the early 1990’s when we banned 2,4,5-T and put in place regulations to ensure our other herbicides had low dioxin levels.

But according last night’s Four Corners program, that regulation has failed us. Before commenting on the program’s claims, I will examine dioxin a bit more.

What is dioxin, and what do we know about its affects?

2,3,7,8-tetrachlorobenzo-p-dioxin, the most toxic dioxin. Ian Musgrave via PubChem

To start with, dioxin is not one thing; it is a mixture of very similar chemical entities of different toxicity and potential harm. The most toxic is 2,3,7,8-tetrachlorobenzo-p-dioxin (TCDD), but any given preparation of herbicide will have varying levels of different dioxins depending on the manufacturing process.

When we evaluate a chemical for potential harm, we look at whether there is a plausible mechanism, whether there are studies in animals and/or tissue culture to support toxicity, and of course epidemiological or other evidence of harm in humans.

There is very clear evidence from animal studies using defined dioxins that chronic exposure to these dioxins can result in wasting disease, kidney damage and a variety of cancers. There is good evidence for a mechanism as well.

Dioxins bind to a protein called the Arylhydrocarbon receptor, which can change the activity of a number of genes that have the potential to lead to cancer.

When it comes to humans things get murkier. As I mentioned, dioxins are a broad range of chemicals with widely varying toxicity. Working out what people have been exposed to, and how much, has been very hard to do in many cases.

Often studies have had to rely on people’s recall of exposure, which may be inaccurate, or measurements made on samples taken many years after exposure. Dioxin is persistent in the body, and it can take up to 7 years to remove half of an absorbed dose, so dioxin can be detected sometime after exposure. However back calculation may be somewhat inaccurate as well.

Also, while in animal studies we can be sure that they have been only exposed to the defined chemical in question, humans may be exposed to a wide range of other chemicals and influences which can confound interpretation.

Many human studies show wildly diverging results on the toxicity of dioxin exposure, so to get a better picture researchers have used meta-analysis, where a large number of studies are grouped together to see if a pattern is revealed.

Using this method, it seems that people who are exposed to dioxins are not more likely to die overall, but are at a slightly increased risk of cancer*, with lung cancers, sarcoma and lymphoma the most prominent cancers involved. There was also an increased risk of respiratory disease and a possibility of decreased immune function.

The Four Corners program covered two broad issues, the effects of past 2,4,5-T exposure in workers employed in weed control, and the levels of dioxin in currently permitted herbicides. I’ll examine the regulation issue.

Dioxins uncovered

As mentioned, in the 1990’s 2,4,5-T was banned in Australia. 2,4-D was permitted, as it has much lower levels of dioxin than 2,4,5-T due to the way it is manufactured. The current legislation requires that the manufacturers have the lowest practical level of dioxin (more on this later).

In 2009 University of Queensland researchers measured the levels of dioxin in a variety of pesticides and herbicides. Some of the pesticides had high levels of dioxin. It appears that the dioxin in these herbicides came from contaminants in a fungicide added to the herbicide. The good news is that the regulator then suspended these pesticides, and set into motion worldwide alert for withdrawal of these herbicides.

They also found more dioxin in 2,4-D than they expected (if you were not paying attention you could have confused the fungicide containing pesticide with 2,4-D). Independent testing showed that several 2,4-D samples had measurable dioxin in them.

In the Four Corners program the terms “low to moderate” in relation to dioxin content of 2.4-D were used, relating this to the FAO, EPA and WHO limits for exposure is, well, impossible. In the paper from the Queensland study their samples of 2,4-D had dioxins equivalent to 0.001-0.2 nanograms of the dioxin TCDD per gram of 2,4-D**. This is less than the international regulatory limit of 10 nanograms of TCDD equivalents per gram of herbicide. To put this in perspective, the highest dioxin levels in 2,4-D were 50-500 times lower than the levels of dioxin in the banned 2,4,5-T.

The “alarming” sample of Chinese sourced herbicide is not given a specific concentration, but is claimed to have 7 times the levels found in the Queensland study. Taking the highest 2,4-D levels in the Queensland study, then this would be 1.4 nanograms TCDD equivalent per gram herbicide, still reasonably below the international regulatory limit.

If workers are using proper safety procedures then they should be safe when using this brand of herbicide, although given recent events this is probably not a safe assumption.

What is disturbing about the whole thing is that the Australian Pesticides and Veterinary Medicines Authority (APVMA) gives only general guidance to pesticide and herbicide producers, that they “.. .both declare the presence of these compounds and to keep levels as low as possible”, presumably assuming that manufacturers will be complying with the international regulatory limits. But they have no way of knowing that.

The APVMA does no testing, and from the looks of things, the herbicide manufacturers do no testing either. While manufacturing processes should result in low dioxin levels, there are a range of things that can go wrong which could result in levels over the regulatory limit (as seen with the pesticides), and if no one is doing any regular testing, we only find out when it is too late.

There is currently a review of 2,4-D underway, reporting later this year, but this issue is larger than 2,4-D. If no one is doing any testing of any of these classes of products for dioxin, then the risks are unknown. Testing and reporting should occur, and should be addressed urgently. The regulator, the APVMA, is in the best position to provide independent and impartial analyses of these compounds.

*what do I mean by slight, well, in this study people who had been exposed to any phenoxy herbicide were 7% more likely to develop cancer than unexposed people, but the confidence interval included the no effect level, so we are only marginally confident that this increase is real. In other studies the effects are a little higher or show no effect. It’s all quite complicated, and a full discussion is beyond the scope of this article.

** Remember there is a mixture of various dioxins with differing toxicities in these products, so everything is normalised to the most toxic dioxin TCDD.