201. Time associations between U.S. birth rates and add-Ons to IVF practice between 2005–2016
- Author
-
Norbert Gleicher, Lyka Mochizuki, and David H. Barad
- Subjects
Adult ,medicine.medical_specialty ,Time Factors ,Databases, Factual ,QH471-489 ,medicine.medical_treatment ,Reproductive medicine ,Fertilization in Vitro ,Birth rate ,Embryo Culture Techniques ,03 medical and health sciences ,0302 clinical medicine ,Endocrinology ,Pregnancy ,medicine ,Humans ,Longitudinal Studies ,030212 general & internal medicine ,Birth Rate ,Preimplantation Diagnosis ,Genetic testing ,030219 obstetrics & reproductive medicine ,In vitro fertilisation ,Assisted reproductive technology ,medicine.diagnostic_test ,business.industry ,Research ,Reproduction ,Obstetrics and Gynecology ,Gynecology and obstetrics ,medicine.disease ,United States ,Reproductive Medicine ,Female age ,RG1-991 ,Female ,business ,Live birth ,Developmental Biology ,Demography - Abstract
Until 2010, the National Assisted Reproductive Technology Surveillance System (NASS) report, published annually by the Center for Disease Control and Prevention (CDC), demonstrated almost constantly improving live birth rates following fresh non-donor (fnd) in vitro fertilization (IVF) cycles. Almost unnoticed by profession and public, by 2016 they, however, reached lows not seen since 1996–1997. We here attempted to understand underlying causes for this decline. This study used publicly available IVF outcome data, reported by the CDC annually under Congressional mandate, involving over 90% of U.S. IVF centers and over 95% of U.S. IVF cycles. Years 2005, 2010, 2015 and 2016 served as index years, representing respectively, 27,047, 30,425, 21,771 and 19,137 live births in fnd IVF cycles. Concomitantly, the study associated timelines for introduction of new add-ons to IVF practice with changes in outcomes of fnd IVF cycles. Median female age remained at 36.0 years during the study period and center participation was surprisingly stable, thereby confirming reasonable phenotype stability. Main outcome measures were associations of specific IVF practice changes with declines in live IVF birth rates. Time associations were observed with increased utilization of “all-freeze” cycles (embryo banking), mild ovarian stimulation protocols, preimplantation genetic testing for aneuploidy (PGT-A) and increasing utilization of elective single embryo transfer (eSET). Among all add-ons, PGT-A, likely, affected fndIVF most profoundly. Though associations cannot denote causation, they can be hypothesis-generating. Here presented time-associations are compelling, though some of observed pregnancy and live birth loss may have been compensated by increases in frozen-thawed cycles and consequential pregnancies and live births not shown here. Pregnancies in frozen-thawed cycles, however, represent additional treatment cycles, time delays and additional costs. IVF live birth rates not seen since 1996–1997, and a likely continuous downward trend in U.S. IVF outcomes, therefore, mandate a reversal of current outcome trends, whatever ultimately the causes.
- Published
- 2021