[tt] Plastic bottle chemical linked to heart disease & type 2 diabetes

[Brian Atkins]

http://www.newscientist.com/channel/health/dn14739-plastic-bottle-chemical-linked-to-heart-disease.html?feedId=online-news_rss20

Exposure to a compound commonly found in plastic food containers is linked to 
heart disease and type 2 diabetes, according to the first large epidemiological 
study in humans.

The findings come just as the controversy surrounding the chemical, bisphenol A, 
is hotting up, with two US government bodies issuing conflicting advice about 
its safety.

BPA is used in the coatings that line food tins, in the hard clear plastics that 
make baby bottles and in dental sealants. More than 2 million tonnes are 
produced worldwide every year, and over 93% of Americans have evidence of 
exposure in their urine.

But most studies investigating potential dangers have been done in animals, 
which metabolise the chemical more slowly than humans, raising questions about 
how reliable such findings are.

In the new study, however, Tamara Galloway at the University of Exeter, UK, and 
her colleagues analysed data from 1455 human adults studied as part of the 
National Health and Nutrition Examination Survey (NHANES) conducted in 2003 and 
2004.

The researchers looked at whether there was an association between urinary BPA 
concentrations and certain diseases.
'Convincing data'

The researchers found that, after adjusting for age and sex, concentrations of 
BPA were higher in people with cardiovascular disease and type 2 diabetes. 
People with the most BPA exposure were three times as likely to have 
cardiovascular disease, and 2.4 times as likely to have diabetes, as those with 
the lowest levels

"This is just a snapshot," says Galloway, who says it gives an indication of 
exposure in the week before the sample was given. The findings cannot prove 
causality. "Now we want to know what happens over time."

Nira Ben-Jonathan, at the University of Cincinnati, who has studied BPA for over 
12 years, was impressed with the paper.

She has looked at how BPA affects human fat tissue, and found that it suppresses 
adiponectin, a hormone that regulates insulin sensitivity. This suppression 
would predispose a person to diabetes. Galloway's research, Ben-Jonathan says, 
complements her own.

Csaba Leranth, at Yale University in New Haven, Connecticut, whose group did the 
first study of BPA in primates, says a human study was much needed. "I find the 
data very convincing," he says.
Conflicting advice

This research will help inform the policy of the US Food and Drug 
Administration, which held a public hearing into the chemical on 16 September, 
at which the paper was presented. Last month, the FDA issued a draft report 
saying current exposures of the chemical posed no danger.

Just a few weeks later, however, the National Toxicology Program, a 
government-funded body charged with providing unbiased, scientifically sound 
evaluations of available evidence, issued its final report on the chemical. This 
concluded that current human exposure levels were cause for concern.

"Changes are happening at doses most humans are exposed to," says John Bucher, 
NTP Associate Director.

Anila Jacob, at the Environmental Working Group, a non-profit organisation based 
in Washington, DC, says the discrepancy between the two bodies is due to how 
they chose to weigh the evidence.

The FDA, she says, relied entirely on three industry-funded studies, because 
they followed certain procedural rules. "The FDA discounted scores of 
independent studies by academic labs," she says. The NTP included these if they 
had merit.

The new paper, says Sarah Vogel, at the Chemical Heritage Foundation in 
Philadelphia, is going to be important because it is confirming effects that are 
seen in the laboratory. "Yes, we can rely on these animal models," she says. "We 
shouldn't be waiting for decades to pass for unequivocal epidemiological studies 
before we act."

Journal reference: Journal of the American Medical Association (vol 300, p 1303)

-- 
Brian Atkins
Singularity Institute for Artificial Intelligence
http://www.singinst.org/