Human Reproduction, Vol. 16, No. 9, 2029,
September 2001
© 2001 European Society of Human Reproduction and Embryology
Letters to the editor |
The effect of intercourse on pregnancy rates during assisted human reproduction
Assisted Conception Unit, Birmingham Women's Hospital, Edgbaston, Birmingham, B15 2TG UK
Dear Sir,
We find the paper by Tremellen et al. (2000) very interesting. It attempts to provide the answer to a question frequently asked by couples undergoing assisted reproduction technology (ART) procedures. This is the first large study to show that sexual intercourse around the time of embryo transfer, far from being harmful, may even be beneficial to embryo development.
The multi-centre study involved couples (women aged 18–40 years) undergoing fresh or frozen embryo transfers. Two hundred frozen embryo transfer cycles from one Australian centre and 400 fresh cycles from two Spanish centres were recruited. A total of 302 couples were randomized to have intercourse around the time of embryo transfer, and 298 to abstain, (654 and 689 embryos transferred respectively). The outcome measures studied were implantation, total and clinical pregnancy rates.
During fresh cycles of IVF or ICSI, the ovaries are usually enlarged and may be very tender around the time of embryo transfer, and most would find sexual intercourse too painful. Questionnaire accounts of compliance amongst the Spanish groups randomized to have sexual intercourse may, therefore, be unreliable. An objective way of ensuring compliance was to ask patients to have sexual intercourse within 2 days before embryo transfer, and perform a microscopic examination of wet endocervical and vaginal smears obtained at the time of embryo transfer for evidence of spermatozoa.
Our second observation concerns how the power calculation was performed by the authors. This study (with a total of 1343 embryos transferred) may have had sufficient power to detect a minimum 50% increase in implantation rate, [i.e. from 7.7–11.5% relative risk (RR) of 1.5, with a cycle cancellation rate of 10%], but does not have the power to detect a RR of 1.5 in pregnancy rate as claimed by the authors. In fact, the authors would have required a total of 4000 cycles of treatment reaching the embryo transfer stage to detect an increase in clinical pregnancy rate from 16.5–19.4% (i.e. RR of 1.21, with an alpha error of 5% and power of 80%) (Dupont and Plummer, 1990
).
Implantation rate, defined as the number of viable embryos at the 6 week scan per 100 transferred embryos is not an ideal outcome measure of an ART programme. The clinical pregnancy, if not live birth rate, is of far more clinical significance (see Instruction to authors, Human Reproduction, Volume 16, 2001).
Conclusions
Whereas this study's findings on the effect of sexual intercourse on implantation rates can be considered valid, a much larger study is required to demonstrate the effect of sexual intercourse on clinical pregnancy rates. It is tempting, but unwise, to assume that the differences seen in clinical pregnancy rates in this study, which involves 600 cycles of treatment, would go on to reach statistical significance with an adequate sample size. This size would however, be very difficult to achieve with a primary study. Perhaps the time has come for a systematic review and meta-analysis of pooled data on the topic.
Notes
1 To whom correspondence should be addressed. E-mail: bolarinde.ola{at}bham-womens.thenhs.com 
References
Dupont, W.D. and Plummer, W.D. (1990) Power and sample size calculations, a review and computer program. Controlled Clinical Trials, 11, 116–128.[Web of Science][Medline]
Tremellen, K.P., Valbuena, D., Landeras, J. et al. (2000) The effect of intercourse on pregnancy rates during assisted human reproduction. Hum. Reprod., 15, 2653–2658.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||