Risks of infertility treatments 'overhyped'

“IVF births carry five times risk of complications,” the Daily Mail reports.

While this headline is essentially true, it is a classic example of a “relative risk” which sounds frightening out of context. In this case, the headline ignores the fact that the number of serious complications found, such as stillbirth and neonatal death was very small.

The story comes from a large study in Australia looking at health outcomes for babies born after fertility treatments (not just IVF), compared with those born after “spontaneous conception”.

Overall, it found that the risk of the following serious complications is around twice as high for babies born after fertility treatment, but the risk is still relatively low:

the rate of stillbirth was 1.1% for any assisted conception compared to 0.5% for spontaneous conception
the rate of premature birth was 7.9% compared to 4.7%
low birth weight 9.4% compared to 4.7%and
neonatal death 0.5% compared to 0.3%

The risk of complications varied according to the method of treatment used, with problems more common in births from conventional IVF (in which sperm are positioned close to an egg, in the laboratory) than from a method called Intracytoplasmic sperm injection (ICSI), in which a single sperm is injected into an egg.

When frozen embryos were used, the higher risk of complications associated with ICSI (but not with standard IVF) were eliminated.

The study also found that babies born to women who had previous fertility problems but no recorded treatment, had the highest risk of complications.

However, the cause of this higher risk is uncertain. It is possible that many of the complications were due to underlying medical problems associated with infertility, rather than the infertility treatment itself.

Where did the story come from?

The study was carried out by researchers from the University of Melbourne and the University of Adelaide in Australia, and was funded by the National Health and Medical Research Council and the Australian Research Council.

The study was published in the peer-reviewed journal PLoS One. PLoS One is an open access journal so the study is free to read online or download as a PDF.

The Daily Mail’s headline was somewhat alarmist and would have better served readers by placing the increase in risk in its proper context. A five-fold increase in a very small risk for serious complications such as still birth is still, essentially, a small risk. However, the paper did include comments from independent UK experts who put the risk in an appropriate context later in their article.

The Guardian carried an accurate and detailed report of the topic.

What kind of research was this?

This was an analysis of the outcomes of more than 300,000 births in South Australia over a 17-year period, including 4,300 births from assisted reproduction.

Researchers compared adverse birth events including stillbirths, premature birth, low birthweight and neonatal death (in which the baby dies within weeks of birth), after spontaneous conception and after fertility treatment.

All available fertility treatments were studied including IVF, ICSI, ovulation induction by drugs, and freezing of embryos.

The researchers make the case that this is the first large scale study to examine the association between different fertility treatments and other complications.

They also analysed outcomes for both single and twin births, excluding higher multiple births (triplets or more).

What did the research involve?

The researchers created a database linking all patients in the state of South Australia who received fertility treatment between January 1986 and December 2002 with the state’s records of all live births and stillbirths in the same period.

Over 20,000 births were excluded from the research. Most of these were pregnancies among mothers under 20 years of age (as only two of these pregnancies were conceived with infertility treatment). Triplet and quadruplet births were also excluded, as were births of babies of indeterminate or unknown sex.

The resulting database included information on 327,378 registered births, 321,210 of which followed a natural conception.

The spontaneous conception group were further classified into:

births to women with no recorded history of infertility in their records and no infertility treatment
births to women who had a recorded diagnosis of infertility but no associated specialist treatment
births as a result of spontaneous conception in women with a previous birth from fertility treatment

They looked at the following types of fertility treatments:

egg donation
gamete intrafallopian transfer (GIFT), where eggs and sperm are placed in the fallopian tubes
intrauterine insemination (IUI), where healthy sperm are placed in the woman’s womb
IVF with fresh embryos
IVF with frozen embryos
ICSI with fresh embryos
ICSI with frozen embryos
minimal medical intervention and ovulation induction (OI) only (ovulation is stimulated using drugs, such as clomiphene citrate)

They analysed and compared the following birth outcomes:

stillbirth
birth weight
low birthweight
very low birthweight
premature birth (before 37 weeks gestation)
very premature birth (before 32 weeks gestation)
late (post-term) birth (after 41 weeks gestation)
small size for gestational age
very small size for gestational age
large size for gestational age
apgar score (this is a quick test of the baby’s health carried out in the minutes after birth)
neonatal death (within a few weeks of birth)

Analyses of birth outcomes other than stillbirth were restricted to live born single babies (296,401) and twins (8,824).

They adjusted all their results for factors which might influence outcomes (confounders) such as maternal age, number of previous births and sex of the baby.

What were the basic results?

The researchers found that compared to birth after spontaneous conceptions, single babies born after assisted conception were more likely to be stillborn (odds ratio (OR) 1.82, 95% confidence interval (CI) 1.34 to 2.48).

Of the babies who survived, those born after assisted conception were more likely to:

be lower birth weight (mean -109g, CI -129 to -89) and very low birth weight (OR = 2.74, CI 2.19 to 3.43)
be very premature (OR = 2.30, CI 1.82 to 2.90)
die in the days after birth (OR = 2.04, CI 1.27 to 3.26)

The outcomes varied by the type of fertility treatment couples had:
Very low and low birth weight, very preterm and preterm birth, and neonatal death were markedly more common in singleton births from IVF and to a lesser degree, in births from ICSI.
When using frozen embryos, all significant adverse outcomes associated with ICSI (but not with IVF) were eliminated.
Frozen embryo cycles were also associated with increased risk of macrosomia (excess birth weight) for IVF and ICSI singletons (OR = 1.36, CI 1.02 to 1.82; OR = 1.55, CI 1.05 to 2.28).
In couples with a history of infertility but no treatment who eventually conceived, babies were nine times more likely to have a very low birthweight, seven times more likely to be very premature and almost seven times more likely to die within the first 28 days of birth.

The study also found that compared to babies conceived naturally:

Babies born after egg donation were at increased risk of being born low or very low birthweight, or be very premature.
Babies born after artificial insemination were lighter at birth and more likely to be low or very low birthweight or very small size for gestational age.
Using drugs to induce ovulation was associated with an increased risk of low birthweight and late birth.

What were the researchers’ conclusions?

The researchers conclude that births after assisted conception show “an extensive range of compromised outcomes” that vary according to the type of fertility treatment used. In some types of treatment, the risk was substantially lower when frozen embryos were used, but this is also associated with an increased risk of excess birth weight.

They suggest that embryo freezing may have a “selective effect” with more compromised embryos less likely to survive.

They also suggest that high rates of birth complications among women previously untreated for fertility problems may be associated with the “poorly supervised” use of fertility drug clomiphene citrate. This group of births should be studied further, they argue.

Further research is needed they argue to pinpoint the cause of the higher risks of birth complications associated with fertility treatment. Routine monitoring of couples undergoing such treatment is also recommended.

Conclusion

The study raises concerns about the link between different types of fertility treatment and poorer outcomes at birth. However, it is uncertain whether the higher risk is down to infertility treatment, underlying health problems associated with infertility, or a combination of both. As the authors point out, the study did not control for the underlying cause of fertility problems.

Although the risks appear higher, the overall risk of these problems is still low. And although the study excluded multiple births, it had no information on whether the single babies born after fertility treatment were the result of multiple pregnancies, a known risk factor for adverse outcomes at birth.

As the authors point out, data for assisted conception pregnancies was not available beyond 2002, and improvements in pregnancy outcomes after assisted reproduction have been noted in recent years.

The findings of the study would have been more useful if information on the underlying causes of fertility problems had been gathered, as these may have had a significant effect on outcomes.

This study reinforces the importance of careful monitoring of women undergoing fertility treatments – and it is unclear how the Australian health system compares with UK NHS or private fertility treatment.