How many IVF embryos to transfer?

1482

CLINICAL PREGNANCY OUTCOME FOLLOWING TRANSFER OF VARIABLE NUMBERS OF FRESH OR FROZEN/THAWED EMBRYOS

Caesarean section and operative

vaginal delivery

rates in

Scotland,1980-87. section rate to be similar for the two groups (13-8% in 1980 to 15-5% in 1987 for teaching hospitals, 10-6% to 125% for the other hospitals). However, the decrease in the operative vaginal delivery rate was much more pronounced for the teaching hospitals (20-1% to 14-3%) than for the others (10-7% to 9-3%). The proportion of births in teaching hospitals remained fairly stable over these years, rising from 28-3% in 1980 to 30-6% in 1987. One effect of an increasing caesarean section rate is a rise in the proportion of women who present having had a section previously and are thus at greater risk of such a delivery in the current pregnancy. The increase in this proportion has been comparable to the rise in caesarean section rate, from 4-9% in 1980 to 7.2% in 1987. When seeking an explanation for the rise in caesarean section, the changing proportions of women in risk groups must be considered; for example, the proportion of first births to older women is growing in Scotland.2 The proportion ofprimiparous women without other risk factors delivered by caesarean section has increased from 7-6% to 9-8% but for similar women of parity 1 the increase has only been from 2-1% to 2.9%. A recent report3 has suggested that caesarean section rates in the USA have reached a plateau at 24-7%. There are two reasons for thinking that Scottish rates may continue to increase: the proportion of the obstetric population with higher-risk attributes is continuing to increase and the likelihood of caesarean section within different risk categories is also growing.

*Number of viable fetuses per total number of given as a percentage

embryos

transferred

in

each group

ceasarean

Social Paediatric and Obstetric Research Unit, University of Glasgow, Glasgow G128RZ. UK

A. H. LEYLAND

1. Cole SK. Scottish maternity and neonatal records. In: Chalmers I, McIlwaine GM. Prenatal audit and surveillance. London: Royal College of Obstetricians and Gynaecologists, 1980. 39-51. 2. Leyland AH, Boddy FA. Maternal age and outcome of pregnancy. N Engl JMed 1990; 323: 413-14. 3. Taffel SM, Placek PJ, Moien M. 1988 US cesarean-section rate at 24 7 per 100 births—a plateau? N Engl J Med 1990; 323: 199-200.

How many IVF

embryos to transfer?

SIR,-Dr Waterstone and colleagues (April 20, p 975) have proposed reducing the number of embryos transferred to two even when other embryos are available, citing their favourable results from such an approach. With specific patient choice to cryopreserve one or more embryos we too have transferred two embryos when others were available to have increased this to three or four. Our figures tell a different story. Our patients range widely in age (23-44) and all indications of infertility are covered; patients are included in our programme irrespective of previous number of IVF attempts or apparent embryo quality. As the table shows, not only is the clinical pregnancy rate apparently higher in cases where only two embryos were available for transfer when compared with elective two-embryo transfers, but also the embryonic implantation rate per individual embryo transferred is significantly higher in this former group (p < 005;

Fisher’s exact test). This is the reverse of the pattern reported by Waterstone et al and suggests that their proposals are too simplistic--or that the data are insufficient, as is often true of conclusions drawn from human IVF results. It is possible that, during selection of embryos for cryopreservation, we may be taking embryos most likely to implant and returning only two lesser quality embryos ("negative selection"), consequently lowering the implantation rate in the elective two-embryo transfer group. This may account for the better results with transfer of two thawed embryos ("positively selected"). Waterstone et al do not mention what happened to their surplus embryos when only two were chosen for transfer, but we presume that they did a positive selection on those transferred. Alternatively, our two-embryo only group may constitute a different patient population to theirs, in that we treat many male-factor couples, with whom fertilisation is often poor but embryonic implantation is good. Hence, having only two embryos to transfer in such cases may not have such a negative impact on pregnancy outcome. Whilst we applaud efforts

to cut down on needless embryonic wastage and multiple pregnancies after transfer of high numbers of IVF embryos, we feel it important to treat each couple on a very individual basis with regard to which embryos and how many we should transfer. And this should be based not only on patient age but also on relative "ovarian age", indication for IVF therapy, and total number of embryos available for both transfer and cryopreservation. Guidelines for the number of embryos transferred will ultimately be undermined by improvements in IVF technique, such as partial dissection of the zona pellucida,’ whereby embryonic implantation will be more readily assured.

Reproductive Biology Associates, Atlanta, Georgia 30342, USA

MICHAEL J. TUCKER HILTON I. KORT JOE B. MASSEY CARLENE W. ELSNER DOROTHY E. MITCHELL-LEEF ANDREW A. TOLEDO

1. Tucker MJ, Cohen J, Massey JB, Mayer MP, Wiker SR, Wright G Partial dissection of the zona pellucida of frozen/thawed human embryos may enhance blastocyst hatching, implantation and pregnancy rates. Am J Obstet Gynecol (in press).

Postmenopausal oestrogen and cardioprotection SIR,—To critics of my April 6 viewpoint article (May 11, p 1161-62) I suggest a careful reading of the thorough review by Barrett-Connor and Bush,1 especially the closing paragraph. Over the past decade these two have been instrumental in making us all aware of the cardioprotective potential of oestrogens. My words of caution about a tendency to push hormone-replacement therapy as a universal cardiovascular preventive are well reflected in the