In Vitro Fertilization (IVF)

Our Pregnancy Rates & Outcomes

Ahead of the Centers for Disease Control and Prevention (CDC) and the Society for Assisted Reproductive Technology (SART)’s annual reports, which will include live birth rates but will not be available until 2017, here, we report the 2015 clinical IVF pregnancy rates for CHR.

2015 CHR Patient Age Distribution

As the pie chart demonstrates, CHR’s patient population continues to age. Already the oldest of any IVF center in the U.S. in prior years (based on national IVF outcome reporting), CHR’s patients have further aged between 2014 and 2015, with significantly more than half of all infertility patients being over 41 years old. Though national data on the FOR of patients at different IVF centers are not available, it is reasonable to assume that CHR’s patients are more “complex” than those of most, if not all, other IVF centers in the U.S. This makes all the sense in the world, considering that over 85% of new patients at CHR have previously failed IVF cycles, often at multiple IVF centers.

FSH and AMH in fresh IVF cycles, 2015
Age (years) AMH FSH
  Mean Median Mean Median
< 30 2.28 0.92 25.0 11.8
30-35 2.21 1.51 9.1 7.9
36-37 1.78 0.78 10.4 7.8
38-40 0.90 0.41 14.9 8.9
41-43 0.99 0.65 12.2 9.7
44+ 0.55 0.26 17.5 10.1

CHR’s patients are not only getting older, they also suffer from increasingly poor functional ovarian reserve (FOR), as demonstrated by age-specific values of anti-Müllerian hormone (AMH) and follicle stimulating hormone (FSH), shown in the table above. Even CHR’s youngest patients, the age group under 30 years, have very poor FOR, with median FSH levels of 11.8 mIU/mL and mean levels as high as 25.0mIU/mL and median AMH levels of as low as 0.92ng/mL and mean levels of only 2.28ng/mL. Like advanced age above 40, low functional ovarian reserve (LFOR) defines patients as demonstrating relatively poor prognosis. Our center’s patient population, based on age and LFOR, therefore, very likely, represents the country’s (if not the world’s) poorest-prognosis patient population to be found in any IVF center. Below listed IVF cycle outcomes for 2015, considering this fact, appear that much more remarkable.

A few disclosures are in order before assessing here presented outcome data: We offer these data to inform patients and colleagues, and not as comparisons to other IVF centers. CHR is in full agreement with ASRM and Society for Assisted Reproductive Technologies (SART) policies, which emphasize for patients and physician providers alike, that IVF outcomes between centers cannot be accurately compared because different centers serve very different patient populations. Different centers have very different mixes of good-, intermediate- and poor-prognosis patients. ASRM/SART, therefore, rightly prohibit member-IVF centers from using their data in such comparisons, and strongly encourage the lay public from not doing it either.

As differences in distribution of good- intermediate- and poor prognosis patients ultimately determine cycle outcomes at IVF centers, comparisons between centers will not become possible until these statistical variations can be properly considered in statistical comparisons between centers. In reforming current national reporting systems, ASRM is attempting to reach such a point; however, even with recently introduced reporting reforms, objective comparisons between centers is not yet possible. Here reported data, therefore, should not be used for such a purpose.

As already noted above, CHR’s patient population is likely at the unfavorable prognosis extreme. Since in poor prognosis populations, many patients’ ovaries do not respond to stimulation and/or patients do not reach embryo transfer for other reasons, these patients are not included in here reported outcome numbers. It, therefore, is important to recognize that here reported outcomes only reflect patients who did reach embryo transfer.

In clinical studies of general patient populations, we strongly advocate use of “intent to treat” reporting of outcomes (i.e., outcome per cycle start). However, in very poor prognosis patients who are clearly advised that in every IVF cycle, they are running a considerable risk of not reaching embryo transfer if they do not produce at least one embryo for transfer, such reporting makes little sense. Such patients understand that their even small pregnancy chance depends on availability of at least one embryos for transfer, and usually want to know what then their chances are if they do produce at least one embryo for transfer. CHR developed this information, even based on how many embryos such a poor prognosis patient has available for transfer, and published it in 2015 (Gleicher et a., Fertil Steril 2015;104:1435-1441).

Considering CHR’s very adversely selected patient population, at approximately 20%, the rate of patients not reaching embryo transfer is actually surprisingly low. This rate, however, obviously greatly varies based on age, prevalence and severity of LFOR. The older a patient and the more severe her LFOR, the higher will be her cycle cancellation rate prior to embryo transfer. As we recently reported in the above-cited study, in worst prognosis patients, approximately 35% of cycles are cancelled prior to embryo transfer. If poor prognosis patients want to know what their clinical pregnancy chances might be per “intent to treat” (i.e., with reference point cycle start rather than embryo transfer) at CHR, they, therefore, likely will be between 20-35% lower than the here reported outcomes.

Finally, since live birth data for 2015 will not be known till the end of this year, it is important to recognize that here reported data are only clinical pregnancy rates. While many, especially early, miscarriages occur before a pregnancy is classified as “clinical,” some miscarriages (at our center ca. 15%) occur after such a designation. Live birth rates, therefore, should be assumed to be approximately 15% lower than here reported clinical pregnancy rates.

Fresh IVF Cycle Pregnancy Rates

2015 fresh IVF pregnancy rates
Age (years) Pregnancy rates (%)
< 30 50.0
30-35 39.1
36-37 18.4
38-39 30.0
40 35.3
41 23.1
42 17.4
43 8.3
44+ 5.9

As the table and figure demonstrate, CHR’s clinical pregnancy rates in practically all age groups are surprisingly high, considering how adversely selected CHR’s patients are. As the figure demonstrates, pregnancy rates, as expected, gradually decline with advancing age, though the age group of 36-37 years demonstrates a significant dip in comparison to younger and older ages. This may be a statistical artefact or, more likely, represents the fact that at approximately that age range young women with premature ovarian aging (POA), also called occult primary insufficiency (oPOI) usually run out of redundancy in their FOR and, therefore, become more resistant to treatment.

Especially remarkable are here reported clinical pregnancy rates in older patients above age 40 years, and again in women at age 43 (8.3%) and above 44 years (5.9%). The latter group included patients up to age 49. It is also noteworthy that here presented 2015 data only partially reflect the changes in CHR’s IVF protocols implemented during the year 2015 for older women as well as younger women with POA/oPOI, which entailed switching to “early” egg retrievals at smaller follicle sizes. These changes were based on the recognition by CHR investigators that women with LFOR usually mature their follicles (and oocytes) at enhanced pace (Wu et al., J. Endocrinol 2015;226:167-180). We are, therefore, hopeful that with full implementation of this practice change throughout 2016, we will be able to add a few additional outcome points, especially in older women.

Clinical Pregnancy Rates for Frozen-Thawed (FET) Cycles

2015 FET pregnancy rates

Clinical pregnancy rates for IVF cycles involving frozen-thawed embryos are separated into routine frozen-thawed cycles (FET), cycles involving donor eggs (FETO) and third-party donations of frozen embryos (FETA). Once again, considering CHR’s patient population, it is actually surprising that there are any frozen embryos available for transfer at all. Here reported pregnancy rates in these cycles are surprisingly good; so good, indeed, that especially FETA cycle outcomes, based on small cycle numbers, may be exaggerated.

Egg Donor Recipient Cycles

  Pregnancy rates (%)
Egg Donor-Recipient CYcles 39.3

In fresh donor egg cycles we have seen in the last two years a number of highly significant changes. Likely because of advancing age, a majority of CHR patients have abandoned 2-embryo transfers in favor of elective single embryo transfers (eSET). CHR is strongly supportive of eSET in older patients since we consider twins medically contraindicated in older women, especially above 43-45 years. Pregnancy rates with eSET are, however, significantly lower than with 2-embryo transfers.

In parallel, we see more demand for gender selection in association with egg donation, often from single women, desirous of a female child. Gender selection also lowers pregnancy chances because of the need for additional embryo manipulation and because embryos of the undesired sex are not transferred.

Combined, these two developments have resulted in a decline in pregnancy rates in oocyte recipient cycles over the last two years. That decline is, however, well compensated as the FETO rate of 34.0% in the table above demonstrates, with eSET and FETO cumulatively producing clinical pregnancy rates at least as good as a 2ET transfer produces. It is important to understand that with 2-ET, and without preimplantation genetic screening (PGS) for gender selection, donor cycles at CHR persistently over the last five years have been in the 60-65% range.

PGD & PGS Cycles

  Pregnancy rates (%)
PGD/PGS CYcles 31.0

In contrast to many other IVF centers CHR does not utilize PGS in attempts to improve IVF outcomes. Most of CHR’s PGD/PGS cycles involve either detection of embryos with single gene diseases or gender selection. As already noted, PGD/PGS requires significant additional embryo manipulation and, therefore, to a degree reduces pregnancy chances. Moreover, patients undergoing PGD/PGS at CHR are usually also relatively older. Consequently, the clinical pregnancy rate of 31% has to be considered excellent.


Considering CHR’s extremely adversely selected patient population (over 90% of new patients have failed IVF cycles before coming to CHR, often at multiple centers), 2015 IVF outcomes have to be considered exceptionally excellent. “Fighting for every egg and embryo” is not only a slogan at CHR; it is engrained in the daily practice of the center and reflected in here reported outcomes.

Last Updated: May 9, 2016

Additional Resources

Letter from a Patient Dear Dr. Gleicher, Dr. Kushnir &amp; the staff at CHR, We are so happy to announce the arrival of our


Artificial Intelligence in IVF: Reproductive Medicine in th... AI or not, embryo selection is still a flawed concept A Scientific American blog recently suggested that artificial intelligence (AI)

Get a Second Opinion
second opinion cta

1/3 of women who have been told they need egg donation actually wind up conceiving at the CHR with their own eggs.