Skip navigation

Anti-oestrogen therapy doesn't work in male infertility

Anti-oestrogen therapy has been used for treatment of infertility in men with oligospermia with variable results. A systematic review from McMaster University in Ontario tell us that it doesn't work [1].


Reports were sought on male infertility or oligospermia treated with clomiphene, tamoxifen or other anti-oestrogens. Studies included needed to fulfil the criteria of:

  • Couples with at least 1 year of infertility
  • Male partner oligospermia with normal gonadotrophins
  • Anti-oestrogen treatment for the male partner
  • Control of placebo, no treatment or equivalent
  • Pregnancy rate as an outcome measure
  • Randomised double blind description


Nine studies fulfilled the inclusion criteria. Six used clomiphene and three tamoxifen, and the duration was from six to 14 months.

The figure shows that most studies had pregnancy rates no higher for treated than untreated groups, and the overall odds ratio of 1.6 had a 95% confidence interval that included 1. There was no difference between clomiphene and tamoxifen.

Quality effect

The reviewers examined the effect of trial quality using a validated quality scale. The scale has a range of 1 - 5 for included studies.

Of the nine studies, two had scores of 3 or more indicating adequate randomisation and blinding. For these two the odds ratio was 0.5 (0.2 - 1.5), showing no effect of treatment.

Seven studies had a quality score of 2 or less, showing inadequate or unclear randomisation and blinding. For these studies the odds ratio was 2.6 (1.2 - 5.2), suggesting a positive effect of treatment.


Bandolier has reported previously the observation that inadequate randomisation and blinding can result in over-estimation of the size of a treatment effect. This study demonstrates yet again the importance of looking at the methodological quality of reports, especially, as here, when the size of an effect is not large.

While the overall result is negative, it may have been possible for a few more studies of poor methodological quality to have tipped the balance to suggest a positive effect that just wasn't there. It's all a question of keeping up the standards.

The bottom line, though, is that this is a treatment we can forget about.


  1. KS Khan, S Daya, AR Jadad. The importance of quality of primary studies in producing unbiased systematic reviews. Archives of Internal medicine 1996 156: 661-6.

previous or next story in this issue