Scientists have jumped the gun in saying that environmental pollutants will cause the infertility of the human race. Where is the evidence? asks Gillian Bentley.

The public could hardly be blamed for panicking about the fertility of humankind given recent reports in the press that human sperm counts are dramatically declining and that food supplies are contaminated with oestrogen-mimicking substances that alter the sexual development of our children. But what is the evidence for this alarming hypothesis? Despite all the media hype, the verdict is far from in.

The link between environmental pollutants and human fertility began a few years ago with the publication of a scientific article in the British Medical Journal. This paper, written by a Danish team of researchers headed by Niels Skakkebaek, purported to show a 50 per cent decline in sperm counts in the last 60 or so years. The article went on to blame this decline, as well as increasing rates of reproductive cancers and developmental abnormalities, on various environmental contaminants that mimic the effects of oestrogens (so-called xeno-oestrogens). The fact that this 1992 paper is the foundation for much of the media hype suggests it deserves closer scrutiny.

Sperm counts usually refer to sperm concentration rather than the total numbers of sperm in an ejaculate. A single sperm sample collected from the same man could yield from 50 to 230 million sperm per millilitre, depending on his age, time since last ejaculation, health status, exercise regime etc. Given that most sperm samples tend to cluster at the lower end of the distribution, a single sample collected on any one day is more than likely to have a low value.

³ÉÈËVRÊÓƵ

Few of the original 61 articles examined by Skakkebk's team in their 1992 article would have controlled for such variation by collecting and averaging several different sperm samples from the same donors. This factor alone is likely to contribute significant bias to averaged figures for sperm counts.

Another important criticism levelled against their paper is that the cut-off point for sperm below which men are considered infertile decreased threefold in 1951 as the result of newer studies on male infertility. It is quite likely that the earlier papers examined might have excluded donors who had sperm counts below the higher threshold. This would automatically bias data from earlier years towards a higher average.

³ÉÈËVRÊÓƵ

Most damaging to the doomsday scenario is research that shows that there is substantial geographic variation in sperm counts. In the United States, New Yorkers (for unknown reasons) have among the highest sperm counts, while Texans and Iowans have among the lowest. Men from more industrialised nations tend to have higher sperm counts. The earlier studies with the highest sperm counts in Skakkebaek's analysis were mostly from New York, while the later studies included other areas with lower sperm counts. Mere geographic variation could explain the results that Skakkebaek and his colleagues reported.

But could xeno-oestrogens be responsible for increases in reproductive cancer rates and other abnormalities, such as undescended testicles (cryptorchidism) and changes in normal urethral tract development (hypospadias) as suggested by the doomsayers?

Again, Niels Skakkebaek and his colleague, Richard Sharpe of the MRC's reproductive biology unit, Edinburgh, are the leading exponents for this link. They do note, however, that evidence for an increase in developmental abnormalities is sparse and conflicting; there are very few papers supporting a consistent worldwide trend at all, and the causes of such anomalies are still not clearly understood. Given that this is the case, we should certainly be cautious in claiming that xeno-oestrogens are responsible for such problems.

There is much less doubt that rates of reproductive cancers have been rising in the industrialised nations in past decades. Causes include our western lifestyles, and particularly our diets. The average British intake of fat comprises 40 per cent of the total diet, while fibre is ingested in low quantities. High-fat and low-fibre diets are known to substantially increase the risk for certain cancers. We also tend to live longer and are more likely to develop these diseases at older ages. Westerners are also characterised by high rates of endogenous reproductive steroids - oestrogen, progesteron and testosterone - all of which may increase our susceptibility. But nobody has linked these findings to xeno-oestrogens in our environments.

³ÉÈËVRÊÓƵ

There is little question that some groups of men and women, mostly industrial workers, are routinely exposed to potentially toxic levels of pollutants. But, such problems do not appear to affect the human population at large. In fact, Stephen Safe at Texas A. and M. University has argued convincingly that the effects of xeno-oestrogens may well be cancelled out because of a large number of natural and man-made compounds present in our environment that are anti-oestrogenic.

Despite these cautionary statements, we should not reject the possibility that the growing cocktail of environmental pollutants can threaten our future health and fertility. This applies not just to xeno-oestrogens, but a whole range of environmental hazards. The current lack of convincing data for an impending human disaster should not lull us into a false sense of security. However, as scientists, we also need to be responsible in advocating any radical public measures to avoid potential contamination, or in heightening concerns that may presently be unfounded.

Gillian Bentley is a Royal Society research fellow in the department of biological anthropology, University of Cambridge.