Assisted Reproduction: A Two-Edged Sword

Almost 2 decades ago, just about the time when my fellowship ended and I was starting on the faculty of UCSF, along came ICSI. Short for intracytoplasmic sperm injection, ICSI involves using a single sperm and manually injecting it into an egg. Voila, the egg fertilizes and becomes an embryo, and later, with the help of mother, a real live baby. No God or Darwin needed to decide on the chosen sperm; in one fell swoop, natural selection is replaced by the judgment of a laboratory technician.

How Good is ICSI?

Make no mistake, ICSI is good. It can go where IVF alone fails to go. The fact is that sperm from any organ can be used with ICSI to create a baby. My medical practice relies on it daily to help azoospermic men become biological fathers. ICSI is also popular. In the latest available data, it was used in 66% of 146,693 U.S. IVF cycles in 2010. That’s 96,817 times a year. Pretty popular by any measure.

The Downside of ICSI

Fortunately for science, the center that invented ICSI (accidentally by the way) has done a great job of following up on kids born with the technique. In fact, they were the first to report that children born with ICSI have higher rates of sex chromosomal issues when compared to naturally conceived children. Since then, concerns have been raised that ICSI kids may also be associated with higher rates of developmental delay and birth defects as well.
What has confused much of this story is our inability to decipher whether the added risk is due to the injection procedure itself or whether it is driven by the genetic risk introduced by infertile parents who require it to conceive. It’s a chicken-or-the-egg issue but in different garb.

The Latest and Greatest

The latest entry in this informational fray about ICSI and birth defects was published last week in the New England Journal of Medicine, the world’s most revered medical journal. An Australian study of 308,974 pregnancies and births over 16 years, both assisted and unassisted, found that birth defect rates were almost 2 fold higher in ICSI infants (9.9%) compared to naturally conceived (5.8%) children. The kinds of birth defects examined were not controversial and included heart, spinal, urinary tract, limb and cleft palate abnormalities, among others, but not minor defects unless they were disfiguring. This study was impressive for the following reasons:
1. It was robust. The sheer numbers of births examined was in excess of 300K+.
2. It was uniform. A single population registry available in Australia was used. Birth defects were clearly defined on a national level.
3. It was comprehensive. Detailed information on children was available from pregnancy to a child’s fifth birthday!
4. It was dissectable. Details and outcomes from various assisted reproduction treatments were available for analysis.
5. It was rigorous. State of the art analytic techniques were used to assess the findings.
One question I had when reading the paper was whether the analysis could help us figure out whether the birth defect risk stems from the ICSI technique or incoming parental risk that is brought to bear on offspring. Here are some observations from the paper that address this issue:
1. The rate of birth defects associated with treatments other than IVF and ICSI were not different from natural conception.
2. The risk of birth defects was also increased among infertile women who did NOT use IVF-ICSI.
3. The risk of birth defects was also increased among women who conceived naturally AFTER using assisted reproduction in the past.
4. Determining the role of paternal issues was not possible, as no information on paternal age or semen quality was included in this study.
In the absence of good information about paternal factors, these observations point out that maternal issues or the technique itself may play role in the higher rate of birth defect findings observed with ICSI.

A Changed World View

So, what to think now? Remember there are dozens of papers addressing this issue over the last 15 years. Given the variation in their quality, to date, I have withheld my opinion on the matter. However, with this study the balance has now shifted for me to truly believe that the birth defect rates are higher with IVF-ICSI than with natural conception. Clearly, the “sword” of ICSI cuts both ways.