Useful Information On Bisphenol A

I commend C&EN on the outstanding cover story articles by Stephen Ritter on BPA (C&EN, June 6, page 13). They certainly helped put the issues in perspective. That many cash-register printouts contain BPA was news to me. This mode of contact with BPA suggests that a study of cashiers, who have the greatest exposure, and their health records is long overdue.

By Sandra G. Rosenthal
Erdenheim, Pa.

Thank you for Ritter’s articles on BPA. This is the best material I have found on the Web, in terms of completeness and balance, on this difficult topic. One potentially significant source of ingested BPA that was not mentioned, however, could represent one of the most prevalent sources of continuous low-dose ingested BPA: The standard liner used for water tanks is BPA-based epoxy.

These liners continuously leach BPA into the water they contain. In addition, these epoxy liners break down over time, and the particulates from the disintegrating epoxy also enter the water systems they supply. As a result of the deterioration of these linings, the tanks periodically need to be emptied, scraped, cleaned, and relined with more epoxy.

Visualize these water tanks baking in the hot sunlight and consider the evidence of how heat increases the amount of BPA leached into the contents of cans, bottles, and other plastic containers. This applies to municipal water tanks as well as industrial water tanks, including those used in the food-processing industry.

Everyone drinking, cooking with, and bathing or showering in water from municipal systems is continually ingesting and being bathed in BPA-laced water. Also, all food processed in water from epoxy-lined tanks is being laced with BPA, even if it might not be packed in water and even if it is packaged in BPA-free containers. This makes it almost impossible for most people to avoid ingesting low doses of BPA. I don’t find it at all surprising that BPA “shows up at low levels in the urine of essentially everyone.”

This struck a chord with me because of my past experience as a city official (city council and mayor), during which time our municipal water tank needed to be relined and the only option offered to replace the deteriorated epoxy lining was epoxy. That was many years ago, long before I had any awareness that BPA might present any problems. If you can do anything to help bring this aspect of the BPA issue to light, it would be a very valuable service. And I bring this up even though I’m not hooked up to a municipal water system; I have my own well.

By Carl Karasti
Winton, Minn.

The articles on BPA and a follow-up in Latest News were most welcome and informative. As a health researcher, I would like to comment on this debate. Although BPA toxicity is of concern for human health (hence the voluminous ongoing research), exposures are low as currently measured using urinary biomarkers—the medians in virtually all study populations are around 2–3 µg/L.

Moreover, the range is quite small, and limited variability in exposure poses problems for epidemiologic investigations. First, narrow variability alters statistical power. Next, it is possible that specimen contamination may be responsible for part of low measured levels, although as the article mentions, dietary and other sources may also account for some of the body burden. Adequate precautions to limit or to measure contamination during specimen preparation have been taken in few if any human exposure studies.

Urinary biomarkers overcome some but not all contamination issues, while blood biomarkers of BPA likely represent only contamination. In addition, low measurement levels result in more noise in the exposure variable (by definition a multiple of method noise), which may lead to unusual statistical associations, including inverse relationships if the exposure measure is essentially a small numerator (exposure biomarker) divided by a larger denominator (urinary dilution factor), tantamount to 1/x.

Furthermore, some study designs may not be appropriate to examine a hypothesis about BPA and health; for example, various reports of BPA in relation to health use convenience cross-sectional data sets. Existing knowledge of exposure patterns as well as biomarker pharmacokinetics and consistency over time makes it difficult to comprehend how concurrently measured BPA represents exposure across the latency period of a chronic disease. BPA levels in urine are correlated with numerous other contaminants; therefore, BPA urinary biomarkers may be a surrogate for correlated urinary metabolites and may signify relationships with multiple body contaminants.

A great advance for health research has been to exploit new technology to measure very low levels of exposure, but more attention should be paid to whether such measures are realistic for epidemiology or toxicology.

By Mary S. Wolff
New York City