Experts Discuss Bisphenol A

There have been numerous attempts to estimate daily human exposure levels but each estimate requires some assumptions. To date, the best measure seems to be BPA levels in the blood (not really actual exposure levels however) which can be related to blood levels of BPA in experimental animals. For humans, levels of BPA are variable and are reported to range from about 0.2 to 4.4 ng/ml (part per billion) in fetuses, children and adults. A summary of BPA levels reported in humans can be found in Vandenberg et al., Reproductive Toxicology 24:139-177, 2007.

The important point to remember is that reported levels of BPA in humans are higher than the BPA concentrations shown to cause effects in laboratory experiments and are within an order of magnitude of the levels shown to cause effects in experimental animal models.

We do not know what level is dangerous. Another important point is that we do not know what age is the most sensitive. However, experimental animal models suggest that fetal and early childhood exposure is the most vulnerable to BPA.

The exposure levels are being determined by laboratories at the Centers for Disease Control and Prevention. Some older data are unreliable.

A recent report from this lab stated that free BPA was not detectable in the urine of the average individual and the highest level was a little above one part per billion (PPB). Total BPA (free and inactive metabolite) levels were a little higher but still in the low PPB range. Levels in milk were about threefold higher than in urine.

The debate rages about what levels of PBA are dangerous. In a review of 700 papers, our panel did not find convincing evidence of reproducible, adverse low-dose reproductive effects. All the robust studies--ones that used a wide range of doses, enough animals to detect subtle effects, measured a wide range of reproductive endpoints, were well designed, used appropriate data analysis, and administered BPA by the oral route (how humans are exposed)--found no adverse reproductive effects.

None of these studies found reproductive malformations, cancer, alterations in puberty, infertility, hormonal changes or obesity, for example. While some of these studies were done in industry laboratories, others were done in government laboratories.

The lack of adverse findings in these large studies is seemingly in contrast to the many effects reported at low doses. However, these low dose effects are often not replicated and are from small studies--some of which injected BPA into a pregnant rat, injected rat pups or, in some cases, injected BPA directly into the brain or spinal cord.

It remains to be determined if any of these low dose effects can be replicated using the relevant route of exposure (oral) in a well-designed study and linked to any adverse reproductive effects.