Pregnancy and child

IVF and risk of stillbirth

“Women who conceive using fertility treatment run a fourfold higher risk of a stillbirth,” the_ Daily Mail_ reported. The newspaper said researchers have warned that women who conceive using in vitro fertilisation (IVF) or intracytoplasmic sperm injection (ICSI) were at increased risk compared to women getting pregnant naturally or using other assisted reproductive treatments.

This study found that the risk of stillbirth was four times greater for these women than those not receiving these IVF treatments. However, the newspaper does not sufficiently emphasise that the absolute risk was still small. Overall, the absolute risk for both naturally and non-IVF assisted conceptions was 4.3 stillbirths per 1000 pregnancies. The risk with IVF and ICSI was 16.2 per 1000, meaning that these techniques raised the risk by 11.9 per 1000 or about 1%. As such, one extra woman in a hundred may experience a stillbirth following these techniques who otherwise would not have done.

The findings are confounded by the fact that women receiving IVF or ICSI may be at increased risk of stillbirths due to factors such as their age. The researchers attempted to adjust for some of these factors, but unmeasured variables cannot be ruled out.

Overall, on its own this study does not completely settle the uncertainty as to whether IVF or ICSI poses an increased risk of stillbirth. What is known is that multiple birth pregnancies are the single biggest risk of fertility treatment and, as the researchers say, twins face an increased risk of preterm birth, low birthweight, and serious health problems.

Where possible, couples undergoing IVF should be encouraged to opt for single embryo transfer in order to reduce the risk of multiple birth pregnancy.

Where did the story come from?

This research was conducted by Dr Kirsten Wisborg and colleagues from the Perinatal Epidemiology Research Unit at Aarhus University Hospital in Denmark, and supported by grants from the Dagmar Marshall’s Fund. The study was published in the peer-reviewed medical journal Human Reproduction.

The Daily Mail and other newspapers added some balance to the debate by quoting the authors of similar studies in the field. One larger study in more than 27,000 women having IVF in Sweden compared outcomes from pregnancies after IVF/ICSI with a control group of 2,603,601 spontaneous pregnancies. In contrast to today’s study, no increase in stillbirths from IVF/ICSI treatment in this larger group was found.

What kind of research was this?

This research used data from a prospective cohort study called the Aarhus Birth Cohort. The analysis of the study data compared the risk of stillbirth in women who had conceived for the first time after fertility treatment, subfertile women (who conceive after a year of trying), and fertile women. To be included, the women had to have had a singleton pregnancy (pregnant with a one baby). The fertility treatment included in-vitro fertilisation (IVF), intracytoplasmic sperm injection (ICSI) and this was compared with non-IVF assisted reproductive technology (ART).

The Aarhus Birth Cohort routinely included information on all single baby pregnancies and births in the town of Aarhus in Denmark from 1989 to 2006. In this data set of over 20,000 pregnancies, the researchers were able to count and compare the number of stillbirths that occurred in women using a variety of non-IVF ART.

One aspect of this analysis showed that women who conceived after IVF/ICSI had higher stillbirth rates compared to those in women who conceived after non-IVF ART. This association was present after other factors thought to also influence the risk of stillbirth were taken into account. However, women with assisted pregnancies do differ from other women in ways that affect risk of stillbirth, for example in the number of children they have already had, age, other diseases and smoking.

Although, the adjustment to the analysis suggests that none of these factors fully explain the results, other unknown factors might and these cannot be ruled out. In addition, other studies have had contradictory results to these, suggesting more research is needed.

What did the research involve?

The researchers explain that it is not known if babies (singletons) conceived through ART have a higher risk of being stillborn as previous research has not fully accounted for important factors that can potentially influence the risk. They say it is particularly important to establish whether it is the fertility treatment, the possible reproductive pathology (why couples are having problems conceiving) of the infertile couples who take it, or other characteristics related to being subfertile that explain any link.

For example, multiple pregnancy (twins and triplets) is an established risk. Although they chose to study only singletons to avoid this possibility, but the researchers also acknowledge that up to 10% of IVF single baby deliveries are the result of twin pregnancies in which one early embryo may have failed to develop.

In this study in Aarhus from 1989 to 2006, women booked for delivery, and who agreed to participate (75% of those asked), completed two questionnaires before the first routine antenatal care visit at 16 weeks into the pregnancy. Together, the two questionnaires collected data on medical and obstetric history, waiting time to pregnancy and fertility treatment, age, smoking habits and alcohol intake during pregnancy, coffee intake, marital status, education and any psychological problems.

The researchers included only women with first time pregnancies and single baby deliveries who filled in the first questionnaire (27,072 women). They excluded 4,268 women with chronic illnesses (such as heart, lung, kidney diseases, diabetes, other metabolic diseases or epilepsy) and 2,638 women with missing information on waiting time to pregnancy and infertility treatment. They analysed the data appropriately using a technique called multivariate logistic regression analysis.

What were the basic results?

From a total 20,166 first-time singleton pregnancies, 82% conceived spontaneously within a year of trying, and 10% conceived after more than a year of trying (classified as sub-fertile). There were 879 pregnancies (4%) as a result of non-IVF fertility treatment and 742 (4%) after IVF/ICSI.

There were a total of 86 stillbirths, making the overall risk of stillbirth, 4.3 stillbirths per 1000 pregnancies. The risk of stillbirth in women who conceived after IVF/ICSI was 16.2 per 1000. The chance of a stillbirth was therefore about four times greater in the IVF/ICSI group after taking into account maternal age, education, marital status, body mass index and intrauterine exposure to tobacco smoke, alcohol and coffee (odds ratio [OR] 4.08, 95% confidence interval [CI] 2.11 to 7.93).

When the researchers did not adjust for any of these factors, the rate was higher (OR 4.44, 95% CI 2.38 to 8.28) showing that these only partially explain the increased risk.

How did the researchers interpret the results?

The researchers say that compared with fertile women, women who conceived by IVF/ICSI had an increased risk of stillbirth that was not explained by confounding factors (other factors they recorded that could have influenced the results).

They say their results indicate that the “increased risk of stillbirth seen after fertility treatment is a result of the fertility treatment or unknown factors pertaining to couples who undergo IVF/ICSI”.

Conclusion

This well-designed prospective study collected a lot of data routinely and followed women through pregnancy until delivery. The researchers note several points of caution:

  • In support of the argument that the increased risk of stillbirth is due to the ART technique and not explained by infertility, the researchers say they found that couples with a waiting time to pregnancy of one year or more and women who conceived after non-IVF ART had a risk of stillbirth similar to that of fertile couples. This may indicate that the increased risk of stillbirth is not explained by infertility.
  • For some confounding factors, the researchers used categories (such as smoking no/yes) instead of counting the number of cigarettes smoked. This may mean that these factors were not fully adjusted for. There may also be unknown reasons for infertility that were not captured in their questionnaire.
  • The ‘vanishing twins’ could have accounted for some of the increase in stillbirths. This is a pregnancy where there were initially two embryos (a twin pregnancy), but as one fails to develop there is just a single baby birth. If these pregnancies took on the risk of multiple births, it could be an explanation of the increased risk. However, the researchers say this is probably not the sole contributor to the increased risk of stillbirth in IVF singletons as the number of ‘vanishing twins’ is small.

Overall, on its own this study does not completely settle the uncertainty as to whether IVF or ICSI poses an increased risk of stillbirth. What is known is that multiple pregnancy remains the single biggest risk of fertility treatment and, as the researchers say, twins face an increased risk of preterm birth, low birth weight and serious health problems. Where possible, couples undergoing IVF should be encouraged to opt for single embryo transfer in order to reduce the risk of multiple birth pregnancies.

It should be remembered that despite this study finding the risk of stillbirth is increased fourfold for couples receiving assistance, the overall risk of stillbirth is actually low. This study found the overall risk of stillbirth in women who had not had IVF or ICSI to be 4.3 per 1000 pregnancies.


NHS Attribution