IVF Success Rate Reporting – Are We There Yet?

29 Jan
CGH Embryo vs. Blastocyst vs. Cleaved Embryo
Ask Dr. Sher A Question

Please join me on Wed, February 6 at 8:30pm Eastern for a live webinar where I’ll discuss this topic in detail and take your questions. For more information or to register, click HERE


I have, for many years, taken issue with the way IVF success rates are recorded, verified (or more accurately, not verified) and reported to consumers. I have published many articles on this topic over the past 25 years and have written on it in my book, “In Vitro Fertilization: The A.R.T. of Making Babies”..

The current system is seriously flawed, and prospective IVF patients bear the brunt of its inadequacies. Because current and prospective IVF  patients are earnestly seeking a standard by which to judge  clinics relative to one another, they are left to rely on published success rates which may or may not realistically represent their chances of conceiving at a given clinic.

The fact that the SART/CDC report, among other parameters, presents IVF outcome  in terms of  live births per embryo transfer procedure,  is a good thing. It is especially relevant given the way that the practice of IVF has evolved over the last few years.

To be more specific, expression of outcome as “births per embryo transfer” is more appropriate than other measurements such as “births per egg retrieval” for two reasons: a) the increasing use of  new embryo genetic tests that  far more reliably allow for  embryo selection and, b)  the recent revelation that the transfer of previously vitrified (ultra-rapidly frozen) embryos yields at least the same success as when fresh embryos are transfered.

The expression of IVF  success in terms of “baby rate  per transfer” broken down into sub-categories based age of the egg provider and upon whether regular IVF was  performed using the patients own eggs or donated eggs were used, is thus much more representitive.

While expressing outcome in terms of baby rate per transfer is a step in the right direction, the statistic that would better reflect effectiveness would be “Live Birth Rate per Embryo Transferred.” This is an important distinction.  For example, two clinics could achieve the same number of live births, while one transferred an average of two embryos per transfer, and the other transferred an average of three embryos. Under the current reporting system, the clinics would have identical IVF success rates.

However, if the more accurate standard of “Birth Rate per Embryo Transferred” were applied to both of these clinics, we would see a very different story:

Clinic A Clinic B
No. of Embryo Transfers



Avg. No of Embryos Transferred



Total Embryos Transferred



Total Live Births



Birth Rate per Transfer



Live Birth Rate per Embryo Transferred



As you can see by this example, the success rates for each clinic would look identical at first glance under the current reporting system. This borders on (hopefully unintentional) deception and keeps the incentive in place for clinics to transfer more embryos per cycle in an effort to boost  success rates.  This, in turn, increases the incidence of high-order multiple births, which put both the mother and babies at risk of short and long-term health complications with associated costs.

The simple step of changing the basis of success rate reporting to “Birth Rate per Embryo Transferred” would put all clinics on an even footing and, at the same time, would  remove much of the incentive to transfer larger numbers of embryos at a time in an attempt to boost success rates.

When discussing IVF success rates, it is very important to understand the natural limitations of human reproduction.  In general, humans are not very efficient at reproducing. Even in young, fertile women with fertile male partners, there is only about a 15% of having a baby per month of trying, and an 70%-80% chance within a full year.

This is largely due to the fact that a high percentage of human eggs have chromosomal abnormalities. In fact, even in the most fertile young women, at best, only half of the eggs are likely to be chromosomally normal or “euploid” (meaning they have the correct number of chromosomes). The remainder will be “aneuploid” (meaning they have more or less than the 23 chromosomes that should be present a mature egg). As the woman progresses in age, the percentage of normal eggs decreases progressively to the point that by age 43, less than 10% are likely to be normal/euploid.

Against this background, it should be clear that we can only work with what nature provides us; we can’t expect that chromosomally abnormal (aneuploid) eggs, upon fertilizatioon will turn into healthy, euploid embryos!

As stated above, there are now ways of identifying the embryos most likely to be chromosomally normal.These methods are far more reliable than microscopic assessment. It is true that embryos comprising less than 5 cells or more than 9 cells by day 3 post-fertilization are developing too fast or too slow and are more likely to be aneuploid. It is also a fact that embryos that fail to develop into expanded, cellular blastocysts by day 6 post-fertilization are almost always aneuploid.

But microscopic grading cannot count chromosomes, and so  even waiting until embryos reach the blastocyst stage to transfer them is not totally reliable.

Conversely, newer genetic testing of  embryos using methods sucg as  comparative genomic hybridization (CGH) can identify and select those that have all 46 chromosomes (“competent” embryos) and which are  most likely to develop in to normal babies. The transfer of such “CGH-normal” embryos allows us to significantly surpass IVF success rates per transfer, compared to when untested embryos are transferred.

The chart below compares the “best probable case scenario” for the following three transfer options:

  1. Transfer of a single CGH-normal embryo
  2. Transfer of a single, Day 5-6 expanded/cellular blastocyst
  3. Transfer of a single Day-3 cleaved embryo.

CGH Embryo vs. Blastocyst vs. Cleaved Embryo

When we increase the number of non-CGH tested embryos transferred to two, we see the following pattern:

IVF Success Rates - 2 embryos

Again, these are “best case” scenarios , the goal we as IVF practitioners need to strive to attain.  Most of us will fall short in one or more categories.  These charts don’t account for the many individual variations in ovarian stimulation protocols, or for complications presented by the couple’s diagnosis and circumstances. Nor does it take into account the adverse affect of anatomical and immunologic implantation dysfunction that can interfere with embryo attachment and thus, IVF outcome.

The takeaway here is that although there are things we can do to increase the likelihood of successful IVF, we are bound and constrained by a number of genetic realities that affect the egg, most of which are outside of our control.

Simply stated, for any given age, there is a natural threshold for the number of “normal” (euploid) eggs that any woman can produce. Then, as she ages beyond her mid-thirties, there is an ever increasing potential that declining ovarian reserve will intercede, making it more difficult to ensure optimal egg development.  This in turn increases the likelihood of chromosomal aneuploidy following the hCG trigger.

Thus, any claim of success rates that across the board fall outside of the natural boundaries presented above should be regarded with a high level of skepticism.


  • simone says:

    Thanks Dr. Sherr

  • simone says:

    Hi Dr. Sherry,

    I am 30 and my husband is 32. We have been ttc naturally for 3 yrs without success. We have discovered mfi with low sperm count and have been recommended for ivf and icsi. My egg reserve and quality is reported as optimal also with an ultrasound demonstrating healthy uterus and ovaries. Our clinic’s ongoing pregnancy rate for our age group is 32% however we have been advised that given our case, we should have a 70% chance at success. Given this, would you recommend that we request CGH testing as well as the immune function testing mentioned elsewhere on your site? Although we are told our chances are good we want to ensure all our bases are covered.
    Thanks in advance for your kind attention.

    • Geoffrey Sher says:

      I don’t know about 70% success? However, I don’t think you need routine immune testing unless you have underlying history to indicate such and I would not do routine CGH testing at your age…even with a coexisting male factor.

      Geoff Sher

Leave a Reply

Your email address will not be published.