Getty Images

By Amanda Gardner

SATURDAY, May 5, 2012 (Health.com) — Babies conceived through in vitro fertilization (IVF) have a higher risk of birth defects than those conceived naturally, but the increased risk may stem from the parents rather than the treatment itself, according to a study published today in the New England Journal of Medicine.

In the study, among the largest of its kind to date, researchers analyzed more than 300,000 births in Australia and found the risk of birth defects to be 26% higher with IVF than with natural, or unassisted, conception—a finding consistent with previous research.

But virtually all of the increased risk associated with IVF, which involves fertilizing eggs outside the body and placing one or more embryos in the mother's uterus, could be attributed to the health and demographic profile of the mother, including her age, body mass index, socioeconomic status, and any health conditions (such as diabetes) she may have experienced before or during pregnancy.

The direct influence of IVF on the risk of birth defects appears to be "very modest," says lead researcher Michael J. Davies, Ph.D., an associate professor of obstetrics and gynecology at the University of Adelaide, in Australia.

"It's important to counsel couples regarding the risk of conception [with] in vitro, but, controlling for all confounding factors, there's not really an increased risk for babies conceived with IVF," says George R. Attia, M.D., director of reproductive endocrinology and infertility at the University of Miami Miller School of Medicine, who was not involved in the study.

However, the same may not be true of all infertility treatments. Intracytoplasmic sperm injection (ICSI), another so-called assisted reproductive technology the researchers examined, was associated with a 77% higher risk of birth defects versus unassisted conception. That figure fell to only 57% after the parents were taken into account, suggesting the procedure itself conferred some of the risk.

ICSI involves removing an egg from the body, injecting it with one carefully selected sperm, and implanting the embryo into the mother. Compared to IVF the procedure is "mechanically more invasive, and bypasses a number of selection processes that may otherwise preclude that sperm from fertilization," Davies says.

In all, roughly 18,000 babies in the study were born with one or more birth defects, such as cerebral palsy or heart abnormalities. The rate of defects was about 7% with IVF and about 10% with ICSI, compared to about 6% with unassisted conception.

Although the findings would seem to suggest that IVF is less risky than ICSI, the two procedures aren't entirely comparable. Unlike IVF, ICSI is primarily used when the male has a fertility problem, such as a very low sperm count.

As the study notes, it's possible—as with IVF—that birth defects associated with ICSI may stem from the underlying causes of male infertility, not the treatment. The researchers weren't able to test this theory, since the current study focused on mothers and took into account the father's occupation only.

Next page: New findings should reassure parents

Previous studies have linked IVF to birth defects, as well as to an increased risk of autism and childhood cancer, but it hasn't been clear how much of the risk can be directly attributed to the procedure.

The new findings regarding IVF should be "reassuring for patients," said Glenn Schattman, M.D., president of the Society for Assisted Reproductive Technology, a professional organization for doctors who provide infertility treatments, in a statement.

In another finding, the study confirmed previous research on the practice of freezing and storing embryos for future use, known as cryopreservation. With both IVF and ICSI, babies born from frozen embryos were less likely to have birth defects than those born from fresh embryos, perhaps because the freezing process helps weed out weaker embryos.

The "fittest and healthiest [frozen embryos] tend to survive, thereby reducing the risk during subsequent stages of development," Davies says.

The study had a key shortcoming that will need to be addressed in future research. All of the participants underwent their infertility treatments between 1986 and 2002, so the findings don't reflect new advances in the field that may have affected the risk of birth defects, the authors say.

Davies and his colleagues presented their findings today at the World Congress in Building Consensus in Gynecology, Infertility and Perinatology, in Barcelona, Spain.