We were pleased to see that the paper by Sharif et al.1 (Vol 105 January 1998) confirmed our previous published reports of the effect of FSH and age on the response of ovaries to gonadotro-phin stimulation2. They confirmed the negative effects of age and of FSH on the number of ampoules of gonadotrophin required, the number of oocytes collected and presented some interesting data on fertilisation and pregnancy rates.
Some differences, however, do exist between our papers which require clarification. Firstly, we did not specifically look at the effect of age andlor FSH on the cancellation rate as there were only three cancellations due to failure of ovarian response in the 174 cycles we studied (I −7%)2. This is considerably lower than the 8.1% cancellation rate reported by Sharif el al., despite there being a similar proportion of women aged 40 years or over in both our studies (9.2% and 9.3%) and a generally similar distribution of FSH concentrations. The difference in cancellation rates is difficult to understand, but may reflect differing approaches to gonadotrophin stimulation between our two units.
Secondly, we did not examine fertilisation or pregnancy rates for several reasons; mainly because of the number of confounding variables. Taking adequate accounts of these would have prevented any valid statistical conclusion. Despite the larger number of couples studied by Sharif et al., we are surprised that they consider these sufficient and they do not comment on any conclusions or corrections made when comparing fertilisation rates in their various branded age and FSH subgroups. For example, did they take account of sperm dysfunction which was the indication for IVF in 35% of their patients and may have adversely affected fertilisation rates disparately between their subgroups? Similarly, they do not comment on whether they took account of confounding factors which would influence pregnancy rates; for example the number of embryos transferred and whether the women had hydrosalpinges. It would be inappropriate to comment on the relative influence of age and FSH concentration on in-vitro fertilisation and pregnancy rates, and to counsel couples accordingly, unless adequate account had been taken of all such confounding variables.
Larger studies are clearly needed to clarify the influence of the woman's age and FSH concentration on fertilisation and pregnancy rates, as well as the outcomes which are most important to couples considering IVF treatment: miscarriage and live birth rates.