Chairman of Urology
Weill Cornell Medical Center
Former President, ASRM
CCRM New York
Below we characterize the best data available on whether ICSI drives greater risks to the offspring, but before we drive in, a word about the quality of data we’re dealing with. Most studies here are:
Not Randomized Controlled: Patients who use ICSI are likely different (higher percentage of male factor infertility) those who did not. We don’t know if the difference in populations accounts for differences in outcomes.
Old: Most studies were published before 2005 and may not account for how ICSI is performed, or applied, today.
Do Not Compare Versus IVF: Many studies compare ICSI-born children versus naturally-conceived children. We think the more relevant question is how they compare versus other IVF-born children.
The data on the impact of ICSI, compared to IVF, on birth defects is mixed. One large population study from Australia showed a significantly increased risk of birth defects for populations using ICSI. However, multiple smaller studies showed no difference and two meta-analysis arrived at the same conclusion.
One Swedish study investigated the impact of ICSI vs. IVF on health complications at birth, ranging from cerebral hemorrhage to infant mortality. The study is large, spans multiple centers and is amongst the best we’ve seen on the subject. Investigators found no difference for any complication, except ICSI patients recorded statistically-significant lower rates of respiratory problems.
The data is more consistent when it comes to urogenital issues and sexual health issues faced by male offspring of ICSI: ICSI boys do appear to be at an increased risk of urogenital malformations requiring surgery, low testosterone levels, and possibly low sperm counts.
Again, this may be driven by the ICSI technique itself or the fact more ICSI patients suffered from male factor infertility to begin with.
A number of studies have looked at how offspring born from ICSI (versus IVF without ICSI) compare along the lines of neurological development. While studies point to higher risk of autism, the data is more divided for mental retardation. As for overall school performance, there is no discernable difference. These observations are after correcting for the fact children born from ICSI are more likely to have a father with a male factor fertility issue.
There’s another, more unknown risk: potentially changing the way our genes express themselves. One mouse study (which may not reflect what happens in humans) noted almost 1,000 genes were changed due to the process of ICSI. Many of these genes, 18%, were changed more than 5x. This finding left investigators concerned about the use of ICSI in the absence of absolute necessity, as we’re still early in understanding what the impact of these genetic changes could be.
When patients elect to fertilize their eggs with ICSI, a portion of those eggs will be damaged. Unfortunately, few studies catalogue the issue and those that do are from a single center and decades old.
However, until we see more recent data we’re inclined to believe egg damage rates from ICSI are in the upper-single digits and possibly higher for certain patient sub-types and at certain clinics.
When ICSI improves the likelihood of a live birth (e.g. severe male factor patients, previous fertilization failure), the risk of egg damage is tolerable. But in certain cases (like “poor responders” with few oocytes) where ICSI shows no benefit, the risk of damaging eggs makes ICSI seem less palatable.