Preimplantation cell number may affect birthweight

Cells may vary in numbers per embryo, be uneven-sized or partially fragmented. Dr Lieberman et al. at the Department of Obstetrics, Gynaecology and Reproductive Biology at the Brigham and Women’s Hospital, Harvard Medical School in Boston, USA have measured the quality of human embryos by counting and assessing the properties of their individual cells. In their paper Rate of cell division and weight of neonates following in-vitro fertilization, they compared the weights of 447 singleton births against cell numbers in embryos when transferred. Earlier reports suggest the possibility that such factors may influence birthweight or anomalies in children. This research, published in the journal Reproductive BioMedicine Online, [www.rbmonline.com/Article/2093 – e-pub ahead of print 1 February 2006] indicates that embryos with fewer cells may produce lightweight babies.

All women who conceived by IVF in their Hospital between 1998 and 2001 were included. Those who lost one or more fetuses during pregnancy, had pre-term births and delivered babies with congenital anomalies (3 cases) were excluded. Embryos scored for their cell number, stage of growth and morphology were transferred 3 days post-fertilization. Mean birthweights of babies produced from embryos with 7 or 8 cells of good symmetry and few fragments were compared with those produced from embryos with <7 or >8 cells.

Gestational ages were also assessed, and the data were adjusted according to numbers of replaced embryos, culture media, maternal age and fetal sex. Sub-groups of women were also analysed. 21% of the patients had embryos transferred with <7 cells, 61% with 7–8 cells, and 18% with >8 cells. Relative birthweights were 3388, 3452 and 3550 g respectively, differences between the lowest and highest groups being statistically significant. Each additional embryonic cell added an average of 42.7 g to birthweights. Other variables analysed included the consequences of assisted ‘hatching’ of embryos in particular patients.

Weaknesses in the data included problems in matching particular embryos to specific singleton births when 2–3 embryos were transferred. Better-designed studies are needed, e.g. transferring single embryos, and involving factors such as smoking, hypertension or diabetes, and different culture media.

Nevertheless, these findings are significant because demonstration of links between birth outcomes and embryonic characteristics prior to implantation are rare. In addition, the observation may not just be unique to IVF babies: it may apply to the general population as well but of course one cannot check this since such embryos remain hidden from observation and analysis.