ARE the millions of children conceived by the fertility treatments IVF and ICSI as healthy as those conceived the old-fashioned way? Apart from the increased chance of multiple births and all the associated health risks, the answer seems to be mostly yes, according to the largest review yet.
The review panel found no evidence of any association between reproductive technologies and a child鈥檚 overall health, its development and psychosocial skills, childhood cancers, major malformations and growth abnormalities. 鈥淏y the time they get to eight years of age they really are indistinguishable from children born normally,鈥� says panel member Joe Simpson of Baylor College of Medicine in Houston, Texas.
However, there are problems at birth. Singleton babies conceived by IVF or ICSI are more likely to be born premature and underweight, and to die soon after birth. And the panel decided there was not enough evidence to draw firm conclusions about the risk of certain rare genetic disorders. What鈥檚 more, no studies have looked at children older than 14, so the possibility of subtle long-term health effects cannot be ruled out.
Advertisement
That worries the President鈥檚 Council on Bioethics, which has called for federally funded, long-term research. 鈥淭here is no definitive understanding of the effects of assisted reproductive technologies on the health of children,鈥� says chairman Leon Kass. 鈥淚t鈥檚 absolutely astonishing.鈥�
Based on past form, such funding is unlikely. Instead, the vast majority of studies reviewed by the panel were conducted in Europe. The panel, set up by Kathy Hudson, director of the Genetics and Public Policy Centre at Johns Hopkins University in Washington DC and made up of fertility experts, gynaecologists and paediatricians, initially looked at 2444 research papers before whittling these down to 169 good-quality studies.
Although the overall picture was encouraging, the panel found that singletons conceived by IVF or ICSI are twice as likely to be premature and to die within seven days of birth. They are also nearly three times as likely to be underweight at birth. The panel speculates that this could be because IVF embryos are more likely to share the womb with twins that do not survive to term.
Given that low birthweight is associated with poorer health and mental performance throughout life, it is odd that the studies looking at older children found no differences. The panel were unable to explain this discrepancy when they presented their initial results last week at an American Society for Reproductive Medicine meeting in Philadelphia.
One suggestion is that the parents of IVF babies tend to be richer and better educated, compensating for any deficits. Interestingly, the panel found that IVF twins fare a little better than twins conceived normally.
In other areas, the review panel decided the evidence was inadequate to draw firm conclusions. For instance, there has been concern about the safety of ICSI (where a sperm is injected directly into an egg), which is rapidly becoming more common than IVF. There have been claims that a penile abnormality called hypospadias is more common in ICSI children, but the panel concluded that the jury is still out.
There have also been claims that certain rare diseases caused by defects in genetic imprinting, such as Beckwith-Wiedemann syndrome, are more common in IVF and ICSI children (New 杏吧原创, 14 August, p 11). These disorders occur when genes inherited from the father or mother are not switched on or off in the correct way.
Beckwith-Wiedemann affects only 1 in 15,000 children, so even if it is far more common after IVF or ICSI, as some studies suggest, the risk would remain tiny. And the panel concluded that any link with IVF is only suggestive.
鈥淭he fear of a reduced pregnancy rate did not seem to be justified. For most patients you don鈥檛 need to transfer two embryos鈥�
But recent work by Gianpiero Palermo at Cornell University in New York supports the idea that assisted reproduction can interfere with imprinting. His team examined the activity of four imprinted genes in samples from the placentas of babies conceived normally, by IVF, ICSI or artificial insemination.
They found that the pattern of expression of imprinted genes was sometimes different for IVF and artificial insemination. 鈥淵ou see a pattern that looks like Beckwith-Wiedemann, but not quite,鈥� says Palermo. And in ICSI, the level of expression of all the genes was often abnormally low. But the study was very small and did not look at the health of the babies.
And even if children conceived by assisted reproduction are shown to have a higher risk of imprinting disorders, the technology may not be to blame, just as it may not be the cause of the problems at birth. Instead, any health effects could be linked to the parents鈥� fertility problems (New 杏吧原创, 29 May, p 17).
This is another question likely to remain unanswered in the short term. To resolve this issue, studies would have to compare IVF and ICSI children with those born to couples who failed to conceive for a year (the definition of infertility) but later conceived without any treatment. In practice, though, recruiting such couples is very difficult.
Disentangling the safety issues is going to become even more difficult as new techniques such as pre-implantation genetic diagnosis are adopted. PGD involves removing one or two cells from an eight-cell embryo, which can then be tested for serious genetic diseases such as cystic fibrosis.
More controversially, more and more clinics are advertising PGD screening for chromosomal abnormalities as a way of boosting pregnancy rates (see Graphic). In the future, it might be possible to screen for imprinting abnormalities too, Palermo says.
What evidence there is suggests PGD is safe, but some feel the rush to adopt it is premature. 鈥淚 would be very careful before I took one or two cells from an [embryo] and pronounced it safe without following the child for a decade,鈥� says Kass. Sweden has even banned its use to screen for chromosomal abnormalities.