Reports in the popular press that the human sperm count in New Zealand is in decline is no surprise. Similar findings reported by Carlsen et al1 in the UK over a decade ago were rather more controversial because the underlying reason had not yet been proposed.

It is now accepted that humans are exposed to chemicals that mimic natural sex hormones via food, water, and the environment.2–4 Occupancy of hormone receptors means that these endocrine-disrupting chemicals (EDCs) have effects very similar to their natural hormone counterparts, albeit at orders of magnitude higher doses. For example, bisphenol-A (BPA), a monomer used in lacquers for lining food cans and in polycarbonate plastics, is approximately 4 orders of magnitude less active than 17β-estradiol. Genistein, a soy phytoestrogen has a similar activity.

The important point is that these chemicals are active at hormone levels (i.e. ng/L in biological fluids). Exposure to EDCs is thought to be responsible for a broad array of adverse effects in humans including precocious puberty in girls5–7 and reduced sperm count8,9—there is no reason to believe that human exposure in New Zealand is different to other parts of the world.4

The time of exposure to EDCs is important. Exposure during pregnancy might have irreversible effects on the offspring if exposure coincides with a key point in development whereas the effects of exposure on adults will probably be minor unless they accumulate over a long period. As a means of ensuring that human exposure does not result in pharmacological activity, regulatory authorities either set tolerable daily intakes (TDIs) or ban the use of the chemicals in food and other products that result in unacceptable exposure to vulnerable groups. For example, Canada is planning to ban polycarbonate plastic babies’ bottles.

A TDI is calculated from the no observable adverse effect level (NOAEL) derived from a toxicology study. An uncertainty factor (UF) is applied to compensate for any deficiency in knowledge concerning the accuracy of test results and the difficulty in estimating the health effects in animals studies and extrapolating these to humans; TDI = NOAEL/UF.

For BPA, the determination of the NOAEL utilises a toxicological endpoint from animal experiments. The selection of an appropriate toxicological endpoint is very important if the TDI is to be relevant to the pharmacological effects resulting from human exposure to the chemical. The NOAEL for BPA, based on a three-generation rat study, is 5 mg/kg body weight/day and the UF is 500 resulting in a TDI of 50µg/kg body weight/day.10

Studies using non-hormonal endpoints (e.g. hepatocarcinoma) to determine the NOAEL give similar results (EUFSA, 2006). However it has been reported that precocious puberty in female animals occurs at doses as low as 2 µg/kg body weight/day2 which suggests that the TDI derived from non-hormonal endpoints is far too high. The determination of an appropriate TDI has implications for human exposure and therefore long-term health effects; clearly we should reconsider the TDI for BPA and set a TDI that is more physiologically relevant.

It is also important to recognise that BPA is just one of many EDCs and that their effects are at least additive.