How America became obsessed with BPA
The $4,450 test
On November 16, the Journal Sentinel reporters explained how they had:
“bought canned foods, storage containers, frozen foods and baby bottles at a Wal-Mart in Milwaukee. Some of the items were known to contain bisphenol A, or BPA, such as the Rubbermaid Premier container, which has a No. 7 recycling number on its bottom. Reporters also wanted to test items for which it was unclear if bisphenol A was present, such as the frozen-food containers.
Testing for bisphenol A is costly, so a limited number of items were purchased and sent to a laboratory - 10 items in all. For eight of the items, three of each were purchased so that the laboratory could repeat the tests. One can of Enfamil liquid infant formula was bought, as were four Gerber 2nd Foods Hawaiian Delight desserts, which came in packs of two.
Items were sent to XeroAnalytical LLC, based in Columbia, Mo. This laboratory is run by Frederick vom Saal, a BPA researcher at the University of Missouri.
Vom Saal's laboratory is one of the few that repeatedly have looked for bisphenol A levels leaching from food containers. The laboratory has performed testing for other media organizations, though this was the first time he had done media tests of items that had been placed in a microwave.
Julia Taylor, a researcher in vom Saal's laboratory, performed the experiments.
The paper sought outside experts to vouchsafe the results.
The University of Washington’s Patricia Hunt one of three “independent” experts on BPA to evaluate the tests on BPA migration from the microwaved infant products. Hunt found cause for concern:
“‘This is stuff that shouldn't be in our babies’ and infants’ bodies,’ said Patricia Hunt, a professor at Washington State University who pioneered studies linking BPA to cancer.” (November 16, 2008)
“…Hunt, the Washington State University scientist, called the levels found leaching from the plastic food-storage containers ‘real doozies.’”
Hunt was cited by the paper for her earlier research on BPA:
“1998: Patricia Hunt, a geneticist at Washington State University, notices that control mice had many more defective eggs when stored in polycarbonate cages.”
This finding had caused a storm of interest as one of the implications of linking BPA to induction errors in mouse chromosomes was the possibility of a connection between BPA and aneuploidy – a condition where chromosomes are abnormal. Such chronic low-dose exposure to BPA in humans could have a similar effect, it was hypothesized, and thus have serious implications an impact on miscarriage. In 2008, Scientific American quoted vom Saal saying of Hunt “In the field one thing people say is, ‘Pat does not get it wrong.’”
But a study paid for by the European Union and Food Safety Authority (Eichenlaub-Ritter et al, 2007), was unable to replicate Hunt’s findings. Another EU-funded study (Pacchierotti et al, 2007) administered various oral doses – acute, sub-chronic, and low chronic – of BPA to mice on a daily basis and failed to find any significant effect on chromosome segregation. The researchers concluded that some other factors must have contributed to Hunt et al’s findings rather than accidental BPA exposure. Neither of these studies was reported by the Journal Sentinel even though they failed to replicate Hunt’s findings.
But this is not the only reason why Hunt is a problematic choice to review the vom Saal laboratory results. Hunt and vom Saal are both signatories of the Chapel Hill Consensus, and are the two most prominent scientific voices claiming that BPA is dangerous. They are also co-authors of a commentary in Environmental Health Perspectives (along with other Chapel Hill signatories) criticizing Good Laboratory Practice. Vom Saal has also been a highly public defender and promoter of Hunt’s research.
Given the degree to which the paper pursues conflicts of interest among those claiming BPA is safe, it would seem incumbent on the paper to disclose Hunt and vom Saal’s common bond. It degrades the pursuit of independent scientific verification to allow a comrade in arms against BPA to review another comrade’s work. One might justifiably wonder whether Hunt’s evaluation of vom Saal’s laboratory is influenced by the view view that ‘Fred does not get it wrong.’
The second “independent” scientist called on to evaluate the results from Vom Saal’s lab was Angel Nadal, a physiologist at the Spanish Biomedical Research Network in Diabetes and Associated Metabolic Disorders in Alicante, Spain. Angel is also a signatory on the Chapel Hill Consensus on BPA, and another co-author of the commentary in Environmental Health Perspectives criticizing Good Laboratory Practice. As with Hunt, Nadal found the results from vom Saal’s lab “to be of concern for human health.”
A professor of cancer and cell biology at the University of Cincinnati, Nira Ben-Jonathan has conducted research on BPA exposure and its links to metabolic syndrome, but did not sign on to the Chapel Hill Consensus or the EHP commentary. Ben-Jonathan “was skeptical that such small amounts of BPA could be detected using the laboratory's method,” said the Journal Sentinel.
Ben Jonathan and Nadal were dropped from the next stage of the evaluation process, wherein the paper used the results to calculate how much BPA children of varying ages would consume if they ate the products. These figures – not actually enumerated in the feature – “were examined by three scientists: vom Saal, Hunt and Robert Moore, a senior scientist at the University of Wisconsin-Madison,” and found them correct.
Nevertheless, it is troubling, from both a journalistic and a scientific point of view, that vom Saal is partially evaluating the work of his own laboratory. Equally troubling is Moore’s comment to STATS by email on the evaluation:
“I remember speaking with a reporter from the Milwaukee Journal-Sentinel. I did check their calculations but that didn't involved anything beyond what I learned in high school. The main thing I remember is being surprised that they didn't actually measure BPA concentrations in foods. Instead, what was measured was the BPA concentration in extracts. The reporter said it was all done according to the standard FDA method, but I cautioned her to report the results as surrogate measurements, not as actual concentrations in foods."
The newspaper story that explained the findings (as opposed to the one cited above, which explained how the testing was done) did not observe this distinction. Instead it reported the findings in a way that suggested that the amounts released (after heating would actually be consumed:
“Products marketed for infants or billed as ‘microwave safe’ release toxic doses of the chemical bisphenol A when heated, an analysis by the Journal Sentinel has found… the Journal Sentinel identified several peer-reviewed studies that found harm to animals at levels similar to those detected in the newspaper's tests - in some cases, as low as 25 parts per trillion. Scientists with an expertise in BPA say the findings are cause for concern, especially considering how vulnerable a baby's development is and how even tiny amounts of BPA can trigger cell damage.”
STATS asked Calvin Willhite, lead author of the NSF International evaluation of the safety of BPA to examine the two articles published in the Journal Sentinel on November 16. [See sidebar for Willhite’s comments] he concluded:
“Much of the confusion evident in the Journal Sentinel article appears to stem from the fact that the authors failed to appreciate the differences in route of BPA exposure (ingestion vs. injection) and how the different routes of exposure influence the body’s metabolic detoxification and excretion of this substance. In the absence of understanding the role of biotransformation and elimination of BPA in relation to route of exposure, one can arrive at erroneous conclusions about risk to human health posed by trace levels of any material present in our environment.”
The claims the paper makes – including raising the specter of breast cancer – come from studies of BPA exposure in animals that administered BPA intravenously, said Willhite. The test, the comparison, the alarm is meaningless. Children aren’t injecting BPA into their veins, they are ingesting it in food, and as every regulatory body has noted, the effects produced by ingestion are different to injection; when exposure to a chemical is through ingestion, risk assessment should also be based on ingestion. The paper tries to get around this awkward fact by noting that
“Animals tested were fed BPA through pumps under the skin that regularly administered the chemical. Some critics say that method exaggerates the chemical's effects. But others say it is an acceptable method because newborns are constantly feeding.”
How would readers understand this passage if they knew that “some critics” included the World Health Organization, the Environmental Protection Agency, and the European Food Safety Authority? What editor would allow such a characterization to make it to print if they were truly aware of the degree of opposition within toxicology and pharmacology to assessing the risks of oral ingestion through subcutaneous injection? It skews the entire story. As Willhite notes,
“The various results with BPA studies that used injection are best considered laboratory artifacts and have no relation to a human health risk assessment since a) all human exposure is via ingestion and b) how the body handles BPA by injection or oral routes of administration in the laboratory is so different. One simply cannot extrapolate injection study results in animals directly to humans who receive very small exposures via ingestion. A good example is the reported changes seen in the rodent mammary gland after injection whereas even massive oral doses for the full lifetime showed no changes whatsoever in that same organ.”
Robert Chapin, Chair of the panel convened by Center for the Evaluation of Risks to Human Reproduction (CERHR) to weigh the evidence and come up with a level of scientific concern over BPA agrees. The Journal Sentinel’s “fatal flaw is not understanding and accepting that intravenous exposure gives us erroneously high blood levels,” he said by email.
“When BPA is ingested by mouth, it is absorbed through the intestines and then passes through the liver. Enzymes in the gut and liver break it down into inactive metabolites. If you give the stuff under the skin or into veins, it bypasses the metabolism in gut and liver, and levels of the active compound are much higher in the blood. In fact, they never reach these levels after oral exposure. That's why any route other than the oral route is effectively irrelevant UNLESS you also measure blood levels so you know how much active parent is there.”
Durodié is also critical of the incestuous nature of the testing and evaluation:
“The Sentinel study is quite remarkable (assuming we can aggrandize it with the label study at all) in allowing Frederick vom Saal to act as judge, jury and executioner. Vom Saal ‘oversaw the newspaper’s testing,’ his laboratory conducted the tests and he then gets to be one of the main commentators about them.
In any other situation, a liberal paper such as this would (to coin a pun) ‘smell a rat’. The conflict of interest transparently evident in vom Saal receiving money for work he then comments on, about an issue he has a well-documented interest in, is quite remarkable.”