IVF screening tests examined

Tuesday December 22 2009

An embryo screening technique used in IVF is safe for single pregnancies, according to BBC News. The technique, called pre-implantation genetic diagnosis (PGD), is used to check embryos for potential problems prior to them being implanted in the mother.

This story is based on research that looked at whether there were additional risks to having this type of test compared to standard IVF treatments alone.  PGD tests are not perfect, and very occasionally even screened embryos will turn out to have genetic disorders The researchers found that there was no increased risk for single pregnancies involving one developing foetus. However, multiple pregnancies (twins, triplets and so on) screened with the technique faced a greater risk of low birthweight and perinatal death.

There are a number of limitations to this study, and some of the differences seen may be due to differences between the PGD and non-treatment groups rather than the pre-implantation testing. The researchers also highlight that this was a small study, and that larger follow-up studies are required.

Where did the story come from?

This research was carried out by Dr Inge Liebaers and colleagues at the Universitair Ziekenhuis Brussel, in Belgium. The study was funded by the WFWG foundation and the Schering-Plough/Merck pharmaceutical company. The study was published in the peer-reviewed medical journal, Human Reproduction.

This study was reported accurately by the BBC and The Daily Telegraph. However, the Daily Mail reported that the rate of death around the time of birth was 2.54% for singleton pregnancies rather than the 1.03% found in the study. The figure of 2.54%  actually relates to deaths around the time of birth in multiple pregnancies without PGD. The rate for multiple pregnancies where PGD had taken place was 11.73%.

What kind of research was this?

This research was looking at the risk rates associated with two techniques for checking and screening embryos during in vitro fertilisation (IVF).

One of the techniques assessed was Preimplantation Genetic Diagnosis (PGD), a method for screening embryos for genetic conditions prior to being implanted into the mother during IVF. This technique is used for parents who have a high risk of passing on genetic conditions because they carry mutations that can cause these conditions. Using this technique allows doctors to implant only those embryos that are thought to be unaffected, therefore avoiding the need for further screening during pregnancy and possibly terminating affected foetuses.

The second technique analysed was preimplantation genetic screening (PGS), which is used in couples at risk of passing on chromosomal abnormalities, for example, having too many or too few chromosomes. These abnormalities can reduce the likelihood of an embryo implanting in the womb and successfully being born, and can cause genetic conditions such as Down’s syndrome.

These techniques were researched through a prospective case series looking at the potential risks of PGD and PGS to the foetus during pregnancy and newborn babies. The researchers also compared their data on PGD or PGS with information on embryos that had been conceived using IVF but not PGD or PGS.

What did the research involve?

The researchers followed up the pregnancies and births of 648 couples receiving PGD treatments and 842 couples receiving PGS. Those receiving PGS included women who were over the age of 36 and receiving IVF, couples who were having trouble conceiving due to male fertility problems, and couples who had a history of recurrent miscarriage. The study followed people who had received PGD and PGS between 1992 and 2005.

The participants all underwent a programme of in vitro fertilisation (IVF). The male sperm was injected into the egg using a technique called intracytoplasmic sperm injection (ICSI). Three days after the egg had been fertilised, one to two cells were removed for PGD/PGS, using a biopsy. The IVF specialists could then screen for genetic mutations or abnormalities in the number of chromosomes that the cells contained. On day three or six after fertilisation the embryos were implanted into their mothers.

Throughout their pregnancies, the participants were given questionnaires for themselves and their obstetrician to complete. Parents were also asked to attend an out-patient clinic with their two-month-old baby if possible. Some participants also received prenatal or postnatal genetic tests to confirm the diagnosis made by PGD/PGS at the embryo stage.

The researchers assessed the incidence of several outcomes, including:

  • still birth: where the foetus is more than 20 weeks old,
  • neonatal death: where the infant dies prior to seven days after birth,
  • perinatal birth: a combined measure of still births and neonatal deaths,
  • low birthweight: defined as less than 2.5kg (5.5lbs),
  • premature birth: birth before 37 weeks of gestation,
  • major malformations: problems leading to major functional impairment and/or requiring surgical intervention.

Rates of misdiagnosis were also assessed. Misdiagnosis was defined as a condition being diagnosed either pre- or postnatally after a negative test result from PGD/PGS.

To make a comparison between PGD/PGS and IVF treatment alone the researchers created a control group using records of recipients of ICSI sperm implantation, the technique used on all PGD/PGS subjects prior to testing. These records were collected between 1991 and 1999 from 2,889 recipients of ICSI at a single centre. Using this type of ‘historical’ control group has limitations because techniques used in IVF may have changed over time, which means that variations in ICSI technique may account for the differences seen between the groups rather than the PGS/PGD itself.

What were the basic results?

The researchers found there was no difference between PGS and PGD in terms of outcomes for foetuses or children. On this basis, the researchers grouped all of the PGD/PGS pregnancies and births together into a single treatment group. This treatment group of 581 delivered children was compared with a control group of 2,889 children who had been delivered following ICSI sperm injection alone. The data on this control group came from children delivered between 1991 and 1999.

There was no difference  between the control group and treatment group in terms of length of pregnancy, birth weight or whether the infants had major malformations. The overall rates of major malformations were 2.13 % in the PGD/PGS group and 3.38% in the ICSI group. Babies from multiple pregnancies that had received PGS/PGD were more likely to have a low birthweight than the ICSI alone group (odds ratio [OR] 1.71, 95% confidence interval [CI] 1.21 to 2.39).

For singleton pregnancies, there was no difference between the treatment group and the control group in terms of rate of death at around the time of birth (perinatal death rate).

However, for multiple pregnancies the risk of perinatal death was greater in the treatment group: the death rates for multiple pregnancies was 11.73% in the PGD/PGS group and 2.54% in the ICSI alone group (OR 5.09, 95% CI 2.80 to 9.90). When combining single and multiple pregnancies, the total death rate was 4.46 % in the treatment group and 1.78% in the control group (OR  2.56, 95% CI 1.54 to 4.18).

Only one of the 170 foetuses/children that received PGD had a negative test result disproved by prenatal or postnatal re-testing. Of the 56 PGS children who were re-tested, none had chromosomal abnormalities.

How did the researchers interpret the results?

The researchers concluded that the embryo biopsy required for PGD or PGS does not seem to change the risk of major malformations, nor does it appear to add risks to the health of newborn singleton PGD/PGS children. However, the researchers added that multiple PGD/PGS children were more often premature and of low birthweight, and there was an increased rate of death at around the time of birth. They say that they cannot yet explain this finding.


This study followed up children who had received embryonic screening as part of IVF, assessing its safety by comparing it with data on children that had been born following ICSI sperm injection alone. Although the follow-up of the children in the treatment group may have been thorough, there are several limitations:

  • The study was relatively small and therefore may not detect small differences between groups.
  • The study used prospectively gathered data on people given PGD/PGS between 1992 and 2005. It compared this with a historical dataset of children who had been born following ICSI alone between 1991 and 1999. Although both groups used the technique of ICSI sperm injection, the exact methods used may have been different due to changes in practice over time.
  • The demographics and the medical histories of the parents in the treatment group and the control group may have been different. These differences may have contributed to the differences seen between the groups rather than the influence of PGD/PGS itself.
  • The PGD children contained a larger proportion of girls than boys, and some of the diseases screened for were sex-linked diseases that primarily affect boys.
  • The researchers followed up children until they were two months old. They suggest that longer-term follow-ups should be conducted.
  • The study used data from one centre in Brussels. The exact methods of the procedures used may vary between centres. On this basis, the results may not be representative of what would be seen at other centres.

In light of their findings, the researchers suggest that the misdiagnosis rates of PGD and PGS are acceptably low and that the techniques do not add to the risks of singleton pregnancies. However, multiple pregnancies should be avoided owing to the increased risk of perinatal death. Current guidelines from the National Institute for Health and Clinical Excellence recommend that no more than two embryos are transferred during any one cycle of IVF to reduce the chances of multiple pregnancies.

Analysis by Bazian
Edited by NHS Choices