The Health of IVF Babies: What Do We Know? What Do We Need to Find Out?

In 1978, in the run-up to the birth of Louise Brown, the first IVF baby, the media speculated greatly over whether she would be normal. 'Had there been anything at all wrong with me,' Brown has reflected, 'I think it would have been the end of IVF.'

This memorable quote was given at the beginning Progress Education Trust (PET)'s most recent online event, 'The Health of IVF Babies: What Do We Know? What Do We Need to Find Out?'. The event provided a look at the current knowledge and future questions regarding the health of IVF-conceived babies, bringing together professionals and the general public.

PET's director Sarah Norcross set the scene for the discussion: eight million babies worldwide have been born from IVF, and with one in six couples experiencing infertility, IVF accounts for two to three percent of all live births in the UK.

The practice of IVF has evolved over the years. Multiple births, for example, are now known to be associated with health risks including premature birth and low birthweight, and so steps have been taken to encourage single embryo transfers. But how else should IVF practices change?

The first speaker, Daniel Brison, professor of clinical embryology and stem cell biology at the University of Manchester, addressed what is known about birthweight, early child growth, and other health outcomes in UK children born from IVF.

He started out by stressing that IVF babies are generally healthy, and that the differences he would describe are very small – they can be seen only by looking at averages across large numbers of births.

It is known that IVF babies have altered fetal growth and birthweight. IVF babies born by fresh embryo transfer have, on average, a slightly lower birthweight, but IVF babies born following frozen embryo transfer are, on average, normal weight or above.

Professor Brison's group followed the weight of children over 25 years from 1991 to 2015, finding that, on average, fresh transfer IVF babies continued to lose weight after birth compared to their naturally-conceived peers, but this weight gap closed by the time they reached school age. Evidence from studies of naturally-conceived children suggests that this 'catch-up growth' is associated with a slightly increased risk of cardiovascular and metabolic diseases in later life.

In another study, Professor Brison and colleagues found that the average birthweight of IVF babies increased over time from 1991-2015, but were not able to map this to a particular factor of the IVF process. They concluded that the culture medium used in the IVF procedure has a minor association with success rate, and that blastocyst transfer appears to be associated with preterm birth. They saw a variation in these effects between different clinics.

We need a multidisciplinary approach to explain these phenomena, Professor Brison said. Embryologists, basic scientists, medical experts, child growth experts and statisticians should all be involved.

The second speaker, Professor Anja Bisgaard Pinborg - chief consultant at the Juliane Marie Centre in Copenhagen, Denmark - outlined research that separated adverse health outcomes caused by IVF to those caused by infertility.

Compared to spontaneous conception, subfertile mothers (those who took over one year to become pregnant) have a 35 percent increased risk of preterm birth over fertile women. Yet, mothers who conceive via IVF are at an even higher risk of giving birth preterm, at 55 percent when compared to fertile mothers. Furthermore, mothers who undergo ovarian stimulation have a 45 percent increased risk.

The risk of preterm birth is lower for IVF using cryopreservation, compared to fresh embryo transfer. Professor Pinborg's research also reiterated findings that, on average, fresh embryo transfer has a higher risk of preterm birth, and a low birthweight.

Her research has found no difference in rates of school performance, autism, or risk of psychiatric disorders in children conceived from subfertile parents and/or through IVF. She outlined other studies that had found no increase of cardiac or metabolic risks in a smaller study than Professor Brison's, but noted that other small studies have found such effects, along with higher blood pressure in IVF children.

The third speaker, Dr Carrie Williams, clinical paediatrician and NIHR Research Fellow at University College London's Great Ormond Street Institute of Child Health, addressed whether IVF children have a different risk of cancer.

She outlined several recent studies that investigated this question, including her own research. A study of over 100,000 IVF children born in 1992-2008 found no significant difference in rates of cancer or cancer types, compared to a non-IVF conceived population. There was a small increase in incidence of a rare type of liver tumour and muscle tumours.

Similar results were found in a Nordic study, with a small increased risk of central nervous system tumours and epithelial tumours. A US study of 275,000 children from 2004-2013 also found no overall cancer risk, differences in cancer types, differences by IVF type or infertility cause. A small increased risk of liver tumours was also found in this study.

Other small studies following frozen embryo transfer over 25 years in Denmark and the Netherlands found a small increased risk and no risk, respectively.

Most studies find no overall difference in cancer risk, concluded Dr Williams. Only one type of cancer, the liver tumour hepatoblastoma, was found consistently between two studies, and this is known to be associated with low birthweight. Hepatoblastoma is very rare, and she stressed that the absolute risk of developing this type of tumour or any other type of cancer is very low.

Since most of these studies have been relatively short, Dr Williams still recommended that longer follow-up studies are needed to follow the health of IVF babies further into adulthood.

Dr Sebastiaan Mastenbroek, senior clinical embryologist at Amsterdam University Medical Centre for Reproductive Medicine, the Netherlands, focused his talk on embryo culture medium.

Is it logical to assume that three to five days of in vitro cultivation prior to embryo transfer during the IVF process will have an effect on the child in later life? Yes, said Dr Mastenbroek: many processes occur during preimplantation development, and between fertilisation and implantation, that will never occur again later in life. Epigenetic changes at this time can have a great deal of influence.

A 2010 study of IVF babies created using two different types of culture medium also indicated a difference in birthweight. This effect was investigated by Dr Mastenbroek's research group in a larger randomised trial, demonstrating a difference in both live births and birthweights between different culture media.

Dr Mastenbroek's group subsequently performed an analysis of 15 types of culture medium, finding various differences between them degrees. 'But we don't know, because the media companies don't disclose exactly what the concentrations of the components are in the culture media.' Two very different types of culture media were chosen for the 2016 study. Dr Mastenbroek urged clinicians to start noting which medium is used, as his group found this information was not always recorded in patient files, and to ask patients to consent to participation in long-term follow-up studies.

In PET tradition, audience members were then invited to question the experts. Is preimplantation genetic diagnosis (PGD) more likely to cause issues than IVF? No, said Professor Pinborg.

Is there association between male infertility and poorer health outcomes if using intracytoplasmic sperm injection (ICSI) as well as IVF? Professor Pinborg said there is a slight increase of chromosomal abnormalities in children, and a small study indicates that sons can have lower fertility.

Did the location of an IVF clinic site affect birthweight, such as if the clinic was located in an area of the UK where other health outcomes are below average? Professor Brison said that although his group hadn't looked at precisely this, they had controlled for the socioeconomic deprivation index in studies of birthweight and child growth, and did not see any strong association.

Why are frozen embryos born closer to natural conception weight? Professor Brison said he was initially convinced it was an issue with frozen embryo freezing, due to studies showing a difference in gene expression following thawing. But now he says the evidence indicates that the key factor is the endometrial environment. He cited evidence such as changes in birthweight associated with how much maternal ovaries were stimulated, and the consistent nature of the outcomes from frozen embryo transfers.

Dr Mastenbroek said the recognition that babies born after cryopreservation have a high birthweight represented a shift in thinking. This an effect which embryologists are still seeking to understand.

Professor Pinborg reported studies indicating a common protocol for frozen embryo transfer resulted in a lack of the hormone relaxin in mothers, causing cardiovascular changes, a higher risk of pre-eclampsia, and an over-grown placenta. She predicted we will see more literature on this and wondered if this protocol should be stopped. Professor Brison said these studies were fascinating, but that despite looking specifically for such an effect, his research group did not find an association between type of endometrial preparation and birthweight.

Did the number of days in culture affect the epigenetics of the blastocyst, and eventual health outcomes? Dr Williams said no associations were found, but her study used only a small cohort. Professor Pinborg said that there were almost no blastocyst transfers included in her studies, as it is a newer procedure and outside the date range of the data included.

Was there any difference specifically when using donor eggs and sperm? None of the experts on the speaker panel were aware of any adverse health outcomes.

Are the pattern of low birthweight, catch-up rates and health outcomes the same for IVF and spontaneously conceived babies? Professor Brison said this was a good question, but all data on long term effects of birthweight and catch-up growth is based on normally conceived babies. But we don't know the mechanism, or whether it applies to IVF children. And we won't know for many years.

Norcross said, in conclusion, that research priorites need to be established for the long-term follow up of children born following IVF, and such multi-disciplinary research needs to be properly funded.