Our Toxicity Experiment in West Virginia
Back in 1998, scientists working for Eastman Chemical – part of the corporate empire established by Kodak photography pioneer, George Eastman – decided to study the effects of dosing rats with a poorly understood compound called 4-methylcyclohexane methanol (MCHM).
Understand that they didn’t really have to do it. MCHM was an older material, one of 62,000 industrial compounds grandfathered in with passage of the 1976 Toxic Substances Control Act. Basically that meant that no further testing was required; the assumption was that these chemicals, which apparently hadn’t killed anyone yet, were unlikely to do so.
Still, Eastman was planning to ramp up its MCHM production (it is currently the only US manufacturer) and, for reasons never made fully clear, the company decided to take a closer look. In this it deserves some credit because, as far as I can tell, no one else, including our regulatory agencies, had MCHM on the research agenda.
When you look at the studies – and we can finally do that because Eastman posted them on its website this week – it’s obvious the scientists were focusing on one question: how poisonous is it? It’s a breath-taking realization, really, that MCHM had been on the market for decades at that point and that no one had bothered to find out.
Pause here for the science writer to catch her breath. This same question came up again last week when thousands of gallons of MCHM spilled into West Virginia’s Elk River and from there into water systems serving some 300,000 people. The contamination plume moved fast;the leak from a storage tank at the Freedom Industries chemical complex was about a mile upstream from the intake system for regional water supplies. The chemical company has since filed for bankruptcy; its website doesn’t acknowledge the incident. In fact, although MCHM is a compound used to wash impurities out of coal, Freedom’s page on coal mining chemicals doesn’t list it (or any other product).
Following the spill, West Virginia issued a “do not use” notice to all consumers on water systems fed by the river. It has since lifted that restriction for almost all consumers, barring parts of one county where contamination levels seemed to have spiked back up. Meanwhile, the plume of contaminated water has rippled from the Elk River into the Ohio River; several days ago, the city of Cincinnati announced that it would shut down its intake valves there due to the MCHM threat. The plume is predicted to be in Indiana by Monday. Following both the plume and the story, however, makes you realize that no one really knows what that means in terms of health risks, that everyone is still asking the same question that Eastman Chemical’s researchers had posed some 15 years ago.
I wrote earlier about the dismaying lack of safety data in a post called Chemical Guesswork in West Virginia. And I was not alone. “An unknown threat,” was the phrase in National Geographic’s story. “The colorless liquid has not been studied extensively,” was the understated analysis from Scientific American (which did a good job anyway of piecing together the limited information available anyway).
My favorite quote came from a chemical safety official interviewed by Chemical & Engineering News: “You never want to be in the position of performing a toxicity experiment like this on your own drinking water supply.”
So true – and I am sure the hundreds of thousands of people exposed to tainted water, not to mention all the domestic animals, wildlife, aquatic life and other species would agree. The fact that the government regulators and agencies really had no idea of the water’s safety became extra obvious this week, when West Virginia first lifted its drinking water bans and then hastily announced that perhaps pregnant women shouldn’t drink the water after all. (I’ve linked here to the West Virginia Gazette, whose reporters have done an exceptional job of telling the story). Private physicians remained wary anyway, warning that young children shouldn’t touch the stuff either.
In the flurry of anxious response it became evident that the state had issued that second warning based on the advice of the CDC. You’ll find that information here in an agency statement (many thanks to my fellow Wired blogger, Maryn McKenna, for alerting me to it) which asks the question “What is the acceptable level of MCHM in Drinking Water?”, answers it with a flat “There should be no MCHM in drinking water,” establishes a 1 ppm safety standard anyway, and then almost visibly dithers: “Due to limited availability of data, and out of an abundance of caution, pregnant women may wish to consider an alternative drinking water source until the chemical is at non-detectable levels in the water distribution system.”
Oh and one other thing. The limited data used by CDC? It wasn’t government research on the compound because as we all know that doesn’t exist. It wasn’t based on independent testing. No, the government relied on the 15-year-old studies done by Eastman Chemical. Or to be precise one of the studies, a kind of superficial analysis that infuriated environmental advocacy groups like the Environmental Defense Fund. In fact – again as reported by the Gazette - the agency had so depended on that one study, which was done with “pure MCHM” that its own river testing looked only for that and ignored the six other chemicals found in the messier “crude MCHM” that actually spilled into the river.
Still, those of us desperate for information will accept any data, any data at all. To that end, let’s look at the Eastman studies, shall we?
The first thing you’ll notice is that there is no human toxicity data. These are studies in species ranging from fathead minnows to rabbits. The study that the CDC used to calculate the safe level of MCHM – this one, in fact – involved 95 rats (45 male, 50 female) force-fed pure MCHM in corn oil for four weeks. The poison concentrations ranged from the 200 to 800 milligram/kilogram level. This is considered roughly equivalent to an exposure in the 200 to 800 part per million range but to be consistent about what these tests say, I’m going to mostly stick with mg/kg. In the mid-range (400 mg/kg) the scientists found a small but consistent pattern of liver and kidney damage, which appeared to be slightly worse in females. They calculated at one-fourth the dose, they would see no effect at all – and its this calculation of zero-effect of 100 mg/kg or 100 ppm in rats that CDC used to set a far more conservative level of zero-effect in humans at 1 ppm. But what the CDC advisory doesn’t tell you is that Eastman made at least one important assumption in this study. It assumed that the damage was transient – “the effects were most likely reversible” to quote from the abstract – and it didn’t track the animals long enough to find out.
To be fair, the Eastman chemists were just trying to figure out how poisonous the compound was. They’d run another more lethal experiment using 30 rats (half male, half female) looking at a higher doses. In that study, they divided the animals into three test groups and three doses. Ten of the rats received oil containing 2000 mg/kg of the compound. All were dead in a day. At half that dose, 3 of the 5 male rats and 4 of the 5 female rats died in less than 24 hours. At half that dose (500 mg/kg) the survival rate went up dramatically – just one female had to be euthanized on the second day. Another study at the 500 ppm level recorded zero mortality. The Eastman researchers calculated that the LD50 (Lethal Dose- 50 percent, a standard toxicity measure of the dose that will kill half of a test population) should be set at 825 mg/kg.
On the well-used Hodge and Sterner toxicity scale, this classifies MCHM as “slightly toxic”. To be an “extremely toxic substance” on this scale, the LD50 rating needs to be 1 mg/kg or less (and this class mostly includes the lethal chemistry produced by bacteria and other microbes, such as the botulinum A toxin). For comparison, the poison nicotine has a definitely dangerous LD 50 of 50 mg/kg in rats; standard table salt has not-to-worry-too-much LD50 of 3000 mg/kg. In other words, the toxicity tests told the Eastman scientists that this was not one of the worst compounds out there – but not one of the benign ones either.
And I say not-benign because the tests also showed that at all levels these test animals suffered. Even at the 500 mg/kg level the rats weakened, stumbled, fell. At the higher levels, they collapsed, their coats became matted with urine and blood – or as the Eastman report puts it, their faces, forearms and paws became covered with “red staining.” So while this wasn’t as murderous as nicotine, neither was it a meaningless poison. Further tests established that it could cause serious skin irritation; because of the resulting discomfort, the scientists euthanized the rabbits used in that study.
In other words, “slightly toxic” is a relative term.
Eastman also ran the Ames assay, which is a preliminary screen for cancer-causing potential. The Ames test involves salmonella bacteria which have been slightly genetically modified. If these altered bacteria are then exposed to a mutagenic compound, some of them will mutate, revert back to their original state. The higher the number of revertants, as they’re called, the more mutagenic the compound. As well as screening for mutagenic potential, the Ames test also allows scientists to sift for possible carcinogens. This follows the premise that all carcinogens are mutagens (although the reverse is not true). But the key point from the MCHM results was that they didn’t find any evidence of mutagenic potential; they found that “the article did not cause a positive increase in the number of revertants.”
Reassuring news, you might say, and that would be definitely so if that finding had been verified by, say, anyone else, if someone, besides the company that manufactures the compound, could vouch for its safety. But this one Ames test and the handful of others conducted by Eastman on a couple hundred rats total, a few rabbits, and that handful of tadpoles (which sickened enough to indicate moderate toxicity to fish) are everything we know about MCHM. Let’s appreciate that Eastman made the information public. But let’s also acknowledge that the company’s message that worried water users should “contact a medical professional. Eastman does not provide medical advice to individuals” is not entirely reassuring.
And what about those beleaguered individuals, the West Virginia mother-to-be afraid to drink the water, the American citizens reminded that the the official message is “trust us” even despite the lack of evidence? This is more disheartening, I think. As I wrote earlier, and as the Eastman studies seem to confirm, this could have been a much more poisonous event. We can take some comfort, I suppose, in reports, that only about 32 people sought medical treatment after exposure to the contaminated water and only four were admitted to the hospital. We won’t find any comfort, though, in the prospect that the system will be improved. Our House speaker, Ohio Republican John Boehner, made a point of reminding us that we already have too much regulation.
He didn’t mention the problem of too little knowledge. Where’s the outcry demanding that we actually invest in research that would provide the information we so desperately needed this time – and will undoubtedly need again? Where is the national realization that this is a situation that puts all of us at risk? Instead of congratulating ourselves on that fortunate survival rate, why aren’t we demanding better protection? By that I mean a richer database, a more transparent information system, and, yes, an update of our embarrassingly antiquated regulation of all things poisonous?
Unless this is really what we want: the cheap solution, the get-what-you-pay for ignorance and apathy from the very systems designed to keep us safe. Yes, there’s nothing like realizing that our regulators and our corporations rely on semi-educated guesses, dependence on corporate testing of chemical compounds, and squeaking by on the lucky chance to make me feel safer. And I’m sure the residents of West Virginia – this month’s guinea pigs in our national toxicity experiments- feel safer too.