Science in Society Archive

What's Wrong with Assisted Reproductive Technologies?

Assisted reproductive technologies are associated with a range of birth defects at least partly due to stresses experienced by germ cell and embryo during culture and storage. Dr. Mae-Wan Ho reports

Since the birth of Louise Brown in 1978, more than a million babies have been delivered worldwide with assisted reproductive technologies or ARTs. An estimated one in ten people of reproductive age are infertile in the industrialized countries, and ARTs are now involved in 1% to 3% of annual births.

ARTs include in vitro fertilization (IVF), induced ovulation (IV) and intracytoplasmic sperm injection (ICSI), which, in turn, form the basis of all newer reproductive and related technologies such as cloning, somatic cell nuclear transplant, genetic modification of germ cells, and recently, transformation of stem cells into germ cells.

Simultaneously as the new reproductive technologies are enthusiastically developed and exploited by biotech companies, however, evidence is emerging that ARTs themselves carry risks to the unborn (see Box 1).

Birth defects associated with ARTs identified between 2002 and 2003

A study in the United States [1] reported a 7.3-fold relative increase in incidence (p=0.0021) of urogenital system birth defect during the first trimester of gestation, based on four patients out of 78 children born with the syndrome during a 4-year period. The condition is extremely rare in the general population. ARTs account for approx. 0.7% to 0.8% of total live births in the US. Of these, 95% to 97% involve IVF using fresh or frozen embryos.

IVF was reported to increase risk of retinoblastoma (malignant tumour of the retina) in the Netherlands [2]. Since 1945, incidence of this disease in the Netherlands has been constant at around 1 per 17 000 live births. Every year, 3000 women in the Netherlands undergo IVF, accounting for 1 to 1.5% of children conceived. In the Vrje University Medical where about 95% of all patients with retinoblastoma are treated, the disease was diagnosed in 5 patients born between Nov 2000 and Feb 2002. The relative risk is estimated to be 4.9 to 7.2 fold

In a study funded by the National Institutes of Health in the United States, three (5%) out of 65 children with Beckwith-Wiedemann syndrome, characterized by enlarged tongue and predisposition to rare cancers, were conceived via IVF [3]. Overall, only 0.8% of births in the US occur after IVF.

In Brazil, four children aged from 6 months to 3 years with several types of cancer were born and diagnosed between 1996 and 2000, whose IVF took place in Sao Paulo [4]. These are out of approx 2000 live births during the five-year period as the result of IVF. Another two children with malignant tumors 2 year and 4 months, and 3 year and 2 months of age, were referred to the same service in 1990 and 2001, also resulting from IVF. In the last edition of Cancer Incidence in Five Continents, the annual incidence rate for cancer for children aged 0-4 years in Brazil was 117.5 cases per 1 000 000. Thus, only one case of cancer would be expected among the IVF children.

A large study in Sweden [5] compared the development of neurological problems in 5680 children born after IVF with 11 360 matched controls. For 2060 twins born after IVF, a second set of controls (n=4120), all twins were selected. The data on neurological problems were obtained from the records of the Swedish habilitation centers. The study found that children born after IVF are more likely to need habilitation services than controls, the overall risk (OR) was 1.7. For singletons (single births), the relative risk was 1.4. The most common neurological diagnosis was cerebral palsy, with increase overall risk of 3.7 after IVF, and 2.8 for IVF singletons. Suspected developmental delay was increased four-fold in children born after IVF. In Sweden, 2% of infants born are the result of IVF, and implantation of two embryos is routine.

In Western Australia [6], the rates of major congenital birth defects in children conceived by IVF and intracytoplasmic sperm injection (ICSI) and control cohorts were found to be 8.6%, 9% and 4.2% respectively.

In Northern Finland [7], 304 IVF children born in 1990-1995 were compared with 569 randomly chosen matched controls. IVF twins (n= 102) were compared with matched control twins (n=103) randomly chosen from the Finnish Medical Birth Register; and IVF singletons (n=153) were compared with singleton controls (n=287).

The overall IVF risks for incidences of preterm birth was 5.6, very low birth weight, 6.2, low birth weight, 9.8, neonatal morbidity, 2.4, hospitalization, 3.2. The prevalence of heart malformations was four-fold in the IVF compared to controls. With the exception of heart malformations, most of the risks were attributed to multifetal births after IVF.

Multiple pregnancy emerged as the most important, albeit not the only risk factor. The European Society of Human reproduction and Embryology (ESHRE) reviewed its guidelines on good practices in IVF laboratories in 2002, and recommended aiming for singleton pregnancies [8]. It is claimed that methods of single embryo transfer have been refined to result in acceptable pregnancy rates, made feasible by improved in vitro techniques in identifying good quality embryos.

However, it appears that the in vitro germ cell and embryo culture techniques may be the cause of certain birth defects [9].

Three 2003 studies all report an unexpectedly high incidence of Beckwith Wiedemann syndrome (BWS) in children conceived with ARTs. Patients with BWS have abnormalities at chromosome 11p15 associated with organ overgrowth and abdominal wall defects as well as increased risk of embryonal tumours. Six of 149 cases were reported from a British BWS registry, the same numbers in a French registry and a further 7 in the USA, all in 2003. These frequencies are extraordinarily high for such a rare congenital condition, representing a significant 4.2-fold increase associated with ARTs.

These findings are reminiscent of reports of sporadic cases of the 'genetic imprinting' disorder, Angelman Syndrome, which has also been linked with ARTs. Angelman Syndrome is characterized by severe mental retardation, motor defects, lack of speech and a happy disposition, and is linked with a loss of function of the maternal allele (copy of gene inherited from the mother) of UBE3A on chromsome 15.

About 50 genes are differentially expressed according to their origin in either the oocyte (egg cell) or spermatozoon (sperm cell). These 'imprinted' genes have roles in growth and development and in tumour suppression (for example, retinoblastoma has been reported by Dutch investigators to be more frequent among ARTs children than normal). At the imprinted genes, only one allele is active (maternal or paternal) and the inactive allele is epigenetically (developmentally) marked, by chemical modification of the histone protein bound to DNA, or by adding a methyl group (-CH3) to the base cytosine on the DNA, or both.

Early in development of the fetal germ cells in both sexes, the germ-cell genomes are erased of methylation marks on the imprinted genes. During maturation of sperm and egg cell, however, re-methylation of the imprinting alleles takes place. DNA methylation, almost invariably associated with repression of transcription, targets one of the two parental alleles to silence it. After fertilization, there are further changes in overall genomic methylation in specific cellular lineages of the embryo but importantly, imprinted alleles are protected from these waves of demethylation and remethylation to maintain their proper dosage effects.

Thus, major epigenetic events take place during both germ-cell development and pre-implantation stages when ART procedures are performed, possibly interfering with the proper establishment (in gamete culture) and maintenance (in embryo culture) of genomic imprints.

Both BWS and AS are associated with imprinted gene clusters. In about 50 to 60% of sporadic cases of BWS, and 5 - 10% of the cases of AS, an epigenetic defect is involved rather than a mutation in the gene. In contrast, all cases of AS and 13 of 19 cases of BWS linked to ARTs were due to epigenetic defects, involving loss of methylation in the maternal allele.

Gene expression - and methylation status - in animal embryos is known to be affected by culture conditions. Notable changes in expression of other genes take place in embryos in culture, but effects on imprinted genes are different in character and are unlikely to be reversible adaptations to altered environmental conditions. In some farm animals, embryo culture and cloning technologies carry high risks of neonatal overgrowth, morbidity and mortality, which, in sheep are associated with loss of imprinting at the Igf2r locus [10]. The so-called large offspring syndrome in animals is reminiscent of BWS in man.

Some researchers have made connections between the effects of the manipulation of gametes and embryos in IVF and those resulting from the manipulation of the maternal diet, even briefly, during early pregnancy [11]. Diets both high and low in protein content can have detrimental effects on embryonic and neo-natal development. High protein diets in sheep during the peri-conceptual period have been linked with low embryo survival and high birth weights similar to the large offspring syndrome. Low protein diets during early pregnancy in rats were found to significantly reduce birth weight of pups. Experimental studies in animal models have clearly established links between poor maternal nutrition to altered prenatal growth and adverse outcomes in terms of cardiovascular and metabolic function in adult offspring.

Intriguingly, similar results following embryo culture are now reported: lighter fetal weight in mice and increased post-natal adiposity, together with body weight gain and abnormal organs in adults. Compared with in vivo derived embryos, culture reduced total cell number and inner cell mass (which eventually gives rise to the fetus). Adding granulocyte-macrophage colony stimulating factor appeared to mostly to overcome those effects.

In vitro culture in ruminants is linked with defective placenta formation. In cow, in vitro embryo production can yield fetuses with abnormal allantoic development and failed blood vessel supply to the developing placenta at day 35 of pregnancy. Similar effects were observed in sheep. But oversized fetuses or placentas have also been reported in cows.

Epigenetic modification in ART embryos is thus a component of a broader causal model linking environmental stress factors with disturbances to development though both transcriptional and epigenetic modification of gene expression (see "Diet trumping genes", this series).

In yet another setback to ARTs, scientists at Brown University, Providence, Rhode Island in the United States reported that using frozen embryos in fertility treatments raises the risk of ectopic pregnancy 17 times. They compared 2452 cycles of IVF using fresh embryos with 392 using frozen transfers. Ectopic pregnancies were 1.8% in the former group and 31.8% in the latter.

Ectopic pregnancies are potentially fatal to the embryo, which gets stuck in the fallopian tube. It causes agonizing pain and can be fatal for the woman if not detected early enough and the embryo removed.

When a woman first has IVF, one to three fresh embryos are usually transferred into the uterus. Within the past decade, as increasing numbers of couples are choosing to freeze some of the spare embryos, giving them a chance to try again if the first attempt fails, thus avoiding the painful process of hormone treatment to induce ovulation.

Storage of frozen embryos also allows patients who have to undergo chemotherapy treatment and other women to create fertilized embryos and delay motherhood.

Dr. David Keefe, the lead researcher, expressed surprise at the finding, "We did not expect it to be so high and we obviously need more research." He said. He thought it could be the thawing process that may disrupt the development of the embryo, making it more prone to stick to the fallopian tube.

A total of 250 000 babies have been born through frozen embryo transfer.

Article first published 03/11/03


  1. Wood HM, Trock BJ and Gearhart JP. In vitro fertilization and the cloacal-bladder extrophy-epispadias complex: Is there an association? J Urol. 2003, 169, 1512-5.
  2. Moll AC, Imhof SM, Cruysberg JRM, Schouten-van Meeteren AYN, Boers M and van Leeuwen FE. Incidence of retinoblastoma in children bron after in-vitro fertilization. Lancet 2003, 301, 309-10.
  3. Possible IVF-Birth defect link. JAMA 2002, 288, 2959.
  4. Odone-Filho V, Cristofani LM, Bonassa E , Braga PE, and Elf-Neto J. In vitro fertilization and childhood cancer. Hematology/Oncology 2002, 24, 421-2.
  5. Stromberg B, Dahlquist G, Ericson A, Finnstrom O, Koster M and Stjernqvist K. Neurological sequelae in children born after in-vitro fertilization: a population-based study. Lancet 2002, 3509, 461-65.
  6. Hansen M, Kurinczuk JJ, Bower C and Webb S. The risk of major birth defects after introdytoplasmic sperm injection and in vitro fertilization. N Engl J Med 2002, 346, 725.
  7. Koivurova S, Hartikainen A-L, Gissler M, Hemminki E, Sovio U and Jarvelin M-R. Neonatal outcome and congenital malformations in children born after in-vitro fertilization. Human Reproduction 2002, 17, 1391-8.
  8. Lambert RD. Safety issues in assisted reproduction technology. Human Reproduction 2002, 17, 3011-5.
  9. Gosden R, Trasler J, Lucifero D and Faddy M. Rare congenital disorders, imprinted genes, and assisted reproductive technology. Lancet 2003, 361, 1975-7.
  10. Young LE, Fernandes K, McEvoy TG et al. Epigenetic change in IGF2R is associated with fetal overgrowth after sheep embryo culture. Nat Genet 2001, 27, 153-4.
  11. Thompson JG, Kind, KL, Roberts, CT, Roberson SA and Robinson JS. Epigenetic risks related to assisted reproductive technologies. Short- and long-term consequences for the health of children conceived through assisted reproduction technology: more reason for caution? Human Reproduction 2002, 17, 2783-6.
  12. "Frozen embryos increase risk of ectopic pregnancy" Maxine Frith, The Independent, 15 October, 2003.

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a tripod?