Declining sperm counts

BMJ 1996; 312 doi: http://dx.doi.org/10.1136/bmj.312.7029.457 (Published 24 February 1996) Cite this as: BMJ 1996;312:457
  1. D M de Kretser
  1. Director Institute of Reproduction and Development, Monash University, Melbourne, Victoria 3168, Australia

    Environmental chemicals may be to blame

    The controversy over whether sperm counts have declined over the past 50 years is reopened by two papers in this issue of the BMJ. In a carefully analysed study of cohorts of unselected men in Britain born between 1951 and 1973, Irvine and colleagues demonstrate a progressive decline in sperm concentration and total sperm number per ejaculate of about 2% a year over 11 years.1 A smaller study by Bujan et al finds no change in sperm counts of sperm donors in the Toulouse area of France over a 16 year period.2

    The controversy over sperm counts began with a metaanalysis by Carlsen et al which showed a decline in a sperm concentrations from 113 x 106/ml to 66 x 106/ml between 1940 and 1990.3 These findings were supported by Auger et al in a study of 1351 fertile men in Paris.4 Commentators have criticised both the retrospective design and the mathematical analysis used by Carlsen et al.5 6 Irvine et al have used data from one laboratory and employed appropriate mathematical analysis to reach their conclusion that sperm counts have declined.1 Bujan et al suggest that the difference between their findings and those of Auger et al4 may be related to the differing environmental conditions of rural and urban populations.2

    The reported decrease in sperm concentration may seem difficult to reconcile with the absence of any detectable decrease in male fertility.5 But important impairment of fertility is often not evident until sperm concentrations decline below 5 x 106/ml.7 This is consistent with one hypothesis advanced to explain the decline in sperm count. Sharpe and Skakkebaek proposed that exposure of the fetal testis to oestrogens or oestrogenic compounds decreased the multiplication of Sertoli cells.8 These cells control the inner environment of the seminiferous tubules in which spermatogenesis occurs,9 and there is strong evidence that each Sertoli cell can support only a limited number of germ cells.10 Hence a decrease in the number of Sertoli cells reduces the output of fertile spermatozoa. Our growing knowledge of how the number of Sertoli cells can be manipulated11 may provide clues to how environmental factors affect sperm output.

    Oestrogens and pesticides are implicated

    The concept that exposure of the fetal testis to oestrogens can interfere with adult sperm production is supported by other data. The sons of women who were given diethylstilboestrol in pregnancy between 1945 and 1970 have been found to have decreased sperm counts and semen volume and an increased incidence of cryptorchidism and hypospadias.12 The Danish Environmental Protection Agency has recently released a report raising concern over possible links between environmental chemicals that have oestrogenic effects and the increasing incidence of cryptorchidism, testicular cancer, and declining sperm counts.13

    Another environmental pollutant with the potential to influence testicular function in utero is the main metabolite of DDT, p,p-DDE, which has been shown to act as an antiandrogen.14 Countries such as Brazil and Mexico used nearly 1000 tonnes of DDT in 1992.15 The metabolic products of DDT and related molecules seem to have the capacity to act through oestrogenic or antiandrogenic mechanisms on the developing male reproductive tract.15

    The time delay between exposure to an agent and development of reproductive dysfunction can often pose problems of linkage. Until recently, for example, the cause of Young's syndrome remained obscure. Young's syndrome occurs in men with bronchiectasis and is characterised by azoospermia due to epididymal obstruction.16 No link has been found with mutations of the cystic fibrosis gene.18 The syndrome was particularly prevalent in Britain and Australia between 1960 and 198016 but is now rarely seen.17 In 1993 Hendry et al noted that many men with Young's syndrome had a history of Pink's disease in infancy.18 Pink's disease disappeared around 1960 with the removal of mercury from teething powders, and the disappearance of Young's syndrome some 30 years later raises the strong possibility that exposure to mercury in infancy may have caused both Pink's disease and Young's syndrome through mechanisms yet to be elucidated.

    Such examples, and our growing understanding of the vulnerability of the male reproductive system to environmental factors, highlight the need for vigilance. Regulatory and research agencies need to determine the most appropriate methods of assessing the actions of agents on the reproductive system, and should undertake whatever studies are necessary to confirm or refute the emerging hypotheses. Delay may compromise the fertility of future generations.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.