Many infertile patients are rightfully concerned that the risk of birth defects might be increased among children conceived using fertility treatments, especially in vitro fertilization (IVF). Here are the titles of the first five articles that popped up when I did a Google search yesterday using “fertility treatment birth defect” as the search terms:
“Common fertility treatments raise birth defect risk, study finds”
“Birth-Defect Risk Higher With Fertility Treatments, Study Shows”
“Infertility Treatments May Raise Birth Defect Risk”
“Fertility Treatments May Raise Risk for Birth Defects: Study”
Bad news? Of course. If I were an infertile patient, headlines like these would be enough to make me consider canceling my appointment to the fertility clinic. These news items all refer to a study recently published in the New England Journal of Medicine entitled, “Reproductive Technologies and the Risk of Birth Defects.” (Here is a link to the original research article: http://www.nejm.org/doi/pdf/10.1056/NEJMoa1008095).
The good news? The study actually showed that most fertility treatments do not appear to increase the risk of birth defects. In fact, here is the last paragraph of the second story listed above (from the New York Times):
“We can now state that a cycle of a single fresh embryo transfer with I.V.F. and, if necessary, followed by the transfer of a frozen embryo will result in no significant additional risk above that of a spontaneous conception,” [the lead author of the study] said. (My emphasis added)
So why the inconsistency? There are several studies that show the prevalence of birth defects is higher among children conceived using fertility treatments than in the general population. If you look hard enough, you can even find a couple of papers that implicate clomiphene in birth defects (a scary thought given that this is the most commonly used fertility drug in the US; in 1991, more than 700,000 clomiphene prescriptions were filled, and I’m confident the number is higher now.) The problem with these studies is that it is not appropriate to compare infertile women with women in the general population. Women who conceive without infertility treatment are generally younger and have different socioeconomic, ethnic, and work backgrounds, and infertile women who conceive are more likely to have never had a baby before. One never knows whether it is the fertility treatment or the underlying differences among the infertile women that are responsible for the observed effects. Ideally, we should compare the birth defect rates among children of women who conceive using fertility treatment to those who conceive spontaneously. The problem with such a study is obvious – infertile women don’t often conceive spontaneously, so it is hard to find suitable controls. (And even then, infertile women who conceive spontaneously are probably different from infertile women who conceive using fertility treatment – they tend to be younger, for one thing.)
But the investigators in the study cited above did just that, linking a South Australian registry of over 300,000 births to registries of assisted conception treatment, birth defects, and fertility clinic data. They compared the rates of birth defects in the children of fertile women to those of infertile women who had conceived spontaneously or using a variety of infertility treatments. They also attempted to adjust the risks based on factors thought to be associated with adverse pregnancy outcomes. Here are the factors they accounted for: “parity, fetal sex, year of birth, maternal race or ethnic group, maternal country of birth, maternal conditions in pregnancy (preexisting hypertension, pregnancy-induced hypertension, preexisting diabetes, gestational diabetes, anemia, urinary tract infection, epilepsy, and asthma), maternal smoking during pregnancy, socioeconomic disadvantage on the basis of the postal code of the mother’s residence (according to the Socio-economic Indexes for Areas), and maternal and paternal occupation.”
And here is the conclusion of the study: “The increased risk of birth defects associated with IVF was no longer significant after adjustment for parental factors. The risk of birth defects associated with ICSI remained increased after multivariate adjustment, although the possibility of residual confounding cannot be excluded.”
In other words, the observed increase in birth defects seen after IVF (about 1.5 times the baseline) was due to patient characteristics rather than the IVF procedure itself. Good news for prospective IVF patients! But what about the increased risk associated with intracytoplasmic sperm injection (ICSI)? I think this is most likely due to patient confounding too, as ICSI is frequently done for male factor infertility and in couples where the husband (who is often much older than the wife) has had a vasectomy. Older men are slightly more likely to father children with certain birth defects (which is why the recommended age limit for men to donate sperm is 39). Unfortunately, the investigators had no information about the age of the man in this study, so they couldn’t control for that variable. (And shame on them for not getting this information).
Also, the condition of male infertility may itself be linked with other conditions that might increase the risk of birth defects. One example of this is cystic fibrosis. Men carrying the gene for cystic fibrosis may have an absence of the vas deferens, and these men typically require sperm aspiration with ICSI to father a pregnancy. (In fact, the relationship between the congenital absence of the vas deferens and cystic fibrosis was first recognized only after cystic fibrosis was frequently noted among offspring conceived using IVF after sperm aspiration).
There were a few other findings in this study worth noting:
– The birth defect rate among the general population was almost 6% higher than most people (including most physicians) realize but consistent with other studies. This is your baseline risk and is (mostly) independent of age. I tell patients to expect a 3% major and 3% minor birth defect risk in any birth, regardless of how the child was conceived.
– There was no particular syndrome that stood out among children conceived using IVF or ICSI (this is reassuring that the procedure itself is probably not inducing a birth defect).
– The birth defect rates were the same in children conceived using fresh and previously frozen embryos.
– “Medically supervised ovulation induction” was not associated with an increased birth defect rate, but “clomiphene citrate at home” was associated with a threefold increase in birth defects, even after controlling for other variables. The authors had no explanation for this, and the number of births was small, so it may just be a spurious finding. (I’m not sure what “clomiphene at home” means, anyway).
So there is both good and bad news about birth defects and infertility. The good news is that fertility treatment (except maybe ICSI) does not increase the risk of birth defects. The bad news is that infertile women are more likely to have underlying problems that do increase that risk, regardless of how their child is conceived. I think the bottom line is: Get as healthy as possible before you get pregnant and don’t do ICSI unless your doctor thinks you really need it to achieve fertilization in your IVF cycle. In fact, I think you shouldn’t do any fertility treatment unless there is some reasonable data to support that doing it will improve your chances of conceiving.
And don’t believe everything you read in the news.