Embryo biopsy and maternal and neonatal outcomes following cryopreserved-thawed single embryo transfer.

We report the largest study to date of both maternal and neonatal outcomes of singleton births following cryopreserved-thawed single embryo transfers utilizing PGT (585 biopsy + 2,191 no biopsy, 2,776 total transfer cycles). Our study presents more than 3-fold more embryo transfers with deliveries than recently reported 7 , and adds to the three existing studies that report maternal and/or neonatal outcomes following contemporary practice of PGT in the United States, where cryopreserved embryos were transferred following embryo biopsy 7 , 8 , 9 Our findings are similar to a recent report of contemporary PGT practice, where preeclampsia was not increased significantly when considering only cryopreserved-thawed embryos following PGT (134 biopsy + 124 no biopsy) 8 In addition, both studies found no increase in placental disorders, gestational hypertension, preterm birth, gestational diabetes, or cesarean section delivery with PGT in cryopreserved- thawed cycles. Our findings differ from a study by Makhijani and colleagues that report an increase in hypertensive disorders of pregnancy, mostly non-severe, following trophectoderm biopsy with PGT 7 However, this increase in mostly mild gestational hypertensive disorders was not associated with an increase in preterm delivery. In addition, this study relied on patient report rather than electronic medical record to document obstetric and neonatal outcomes, which is subject to recall bias. Our study used data that linked SART CORS to electronic medical records of birth certificate deliveries and PELL hospitalization data in Massachusetts, so is not subject to recall bias. Makhijani’s study is the only report of a reduced incidence of cesarean section deliveries with PGT, which may have occurred by chance, in contrast to the present study. In addition, our findings differ from a very large national study of perinatal outcomes of cryopreserved-thawed single embryo transfers utilizing the SART CORS database (2014–2015), that reported a slight but significant increase in preterm birth of 1.6% with PGT 9 We did not find this increase with PGT, which may be due to our smaller sample size of neonatal outcomes or to our consideration of a wider number of years in the SART CORS database (2014–2017). The present report is consistent with the study by Li and colleagues in finding no differences in birth weights of infants following PGT 9 . Other published studies of maternal or neonatal outcomes following PGT are difficult to compare to our study, as they utilized cleavage stage biopsy 6 , 10 , 11 , 18 , 19 or polar body cell biopsy 12 The removal of one cell from an 8-cell cleavage-stage embryo or the biopsy of a polar body may have different consequences than the removal of 5 cells from the trophectoderm of a blastocyst. The process of polar body biopsy, in particular, has been associated with low birth weight of singletons and a trend toward singleton preterm delivery 11 , 12 . There may be an increase in low birth weight of twins following cleavage cell biopsy 18 . Some studies also combined fresh transfers with cryopreserved-thawed transfers for analysis, which could lead to confounding for some outcomes such as birthweight and preeclampsia 6 , 10 , 14 , 18 , 19 Our initial finding of increased maternal length of stay following embryo biopsy did not remain significant after adjusting for confounders, particularly for infertility diagnoses. We considered reasons for prolonged maternal stay, finding that maternal bleeding after vaginal or cesarean section delivery was more common after biopsy which may have contributed to the prolonged stay. It is unclear which infertility diagnoses were associated with bleeding and increased stay following biopsy.

Considering the increasing utilization of PGT in ART practice in the United States, it is reassuring that we found no significant maternal or neonatal effects of PGT in this large study considering contemporary practice of embryo cryopreservation utilizing single embryo transfer. Additional prospective studies are still needed to confirm these findings. Future studies should consider the effect of endometrial preparation for cryopreserved- thawed embryo transfers with PGT, as this was reported to be more predictive of hypertensive disorders in pregnancy than the use of trophectoderm biopsy 7 It is possible that the presence of a corpus luteum in natural cycle cryopreserved-thawed embryo transfers may reduce disorders of placentation due the production of relaxin, a vasoactive substance, by the corpus luteum 20 , 21 , 22 . Our study has several strengths as well as a few limitations. Strengths include its relatively large sample size from a single state over several years, which adds to the growing body of safety data regarding PGT for both maternal and neonatal outcomes. We excluded many possible confounders from analysis such as fresh embryo transfers, multiple gestations, and donor egg cycles 14 . We controlled for mother’s age, race/ethnicity, education, parity, body mass index, year of birth, insurance at delivery, and infertility diagnosis which were confounders in the present study. Limitations include the lack of information about the method of fertilization in the SART CORS database for cryopreserved embryo transfer cycles during this time period, and the stage of embryo biopsy which was not known conclusively. It is likely that most, if not all, embryos were biopsied at the blastocyst stage in our study, as this was the standard of care in MA from 2014–2017, but we cannot be certain of this. In addition, the method of endometrial preparation was not included in the SART database. Our IVF clinic was one of six in Massachusetts that provided data to SART during this time period. All endometrial preparations for FET’s at our clinic were programmed hormone replacement (HRT) cycles during this time period. We contacted IVF medical directors at two large IVF clinics in the state, who also reported that they utilized programmed HRT cycles exclusively during this time period. Thus, natural cycle or Letrozole-stimulated FET cycles do not appear to be common in MA at the time of this study or used preferentially with any patient diagnosis. Pregnancy-induced hypertension or preeclampsia in a previous pregnancy was not known in PELL, so could not be controlled for in our study. Lastly, the correctness of our data is subject to correct ICD9/10 coding by physicians.

Data sources for the study were the Assisted Reproductive Technology (ART) surveillance data from the Society for Assisted Reproductive Technology Clinic Outcome Reporting System (SART CORS) and birth certificates and maternal and neonatal hospitalization discharge data in the Massachusetts Pregnancy to Early Life Longitudinal (PELL) data system. SART CORS contains data entered by ART clinics and reported to the Centers for Disease Control and Prevention in compliance with the Fertility Clinic Success Rate and Certification Act of 1992 (Public Law 102–493). The SARTCORS database contains patient demographics, infertility diagnoses, ART cycle-specific, pregnancy, and outcome data. Data are obtained from approximately 90% of ART clinics in the United States and from all Massachusetts clinics. SART CORS data are validated annually with a random sample of clinics having on-site visits for chart review to compare data reported by the clinic to information recorded in patients’ charts. In 2017, most data fields selected for validation were found to have discrepancy rates of ≤ 5%, except for cycle start date and several of the infertility diagnoses 16 The PELL data system links Massachusetts birth certificates and fetal death records to hospital utilization data for the delivery event and subsequent non-delivery hospitalizations for the mother and infant. The data have been linked for 98% of births and fetal deaths for individual women and their children since 1998. PELL data are linked through randomly generated unique ID’s for mothers and infants. Data are maintained by the Massachusetts Department of Public Health (MDPH). SART CORS was linked to PELL as previously described 17 Briefly, SART CORS and PELL were liked for Massachusetts resident women delivering in Massachusetts hospitals for deliveries from 2004–2017 using a deterministic five phase linkage algorithm based on mother’s date of birth, her first and last name, father/partner’s last name, baby’s date of birth, plurality, and infant sex. The linkage rate for 2004–2017 data was 91.5% overall, and 94.9% for deliveries in which the mother’s zip code and the clinic were both located in Massachusetts. Our sample size for this study was determined from the SART CORS database from January 1, 2014 to December 31, 2017 in Massachusetts consisting of a total of 53,979 ART cycles. We focused on these years because they reflect contemporary practice, with the last 3 of these 4 years reporting the most cases, with ability to link to obstetrics and neonatal records. Of these ART cycles, 9,569 were cryopreserved-thawed, autologous embryo transfers, of which 4,693 cycles were single cryopreserved-thawed embryos transferred on day 5, 6, or 7 with biopsy (1,188) or no biopsy (3,505). A total of 2,776 cycles were linked to deliveries from the PELL birth certificate for calendar years 2014–2017 (yielding 585 with biopsy and 2,191 with no biopsy). We compared obstetrical and neonatal outcomes of deliveries having embryo biopsy (n=585) to those having no biopsy (n=2,191). From 2014–2017 in Massachusetts, the standard of care was to perform biopsies of blastocysts on days 5–7 after fertilization for preimplantation genetic testing, with cryopreservation of embryos until biopsy results were available. This was followed by transfer of a single euploid embryo in a subsequent cryopreserved-thawed transfer cycle. We used Chi-square to test for significant differences across groups for categorical and binary variables. We performed logistic regression to calculate odds ratios and 95% confidence intervals for embryo biopsy vs. no biopsy (reference), adjusting for mother’s age, race, education, parity, body mass index (BMI), birth year, insurance, and infertility diagnosis. Maternal ICD-9, ICD-10, and other diagnostic outcomes obtained from PELL included: placental abnormalities [placental abruption (641.2, 762.1, O45), placenta previa (641.1, O44), vasa previa (663.5, O69.4), placenta accreta, increta, orpercreta (667.0, O43.21, O43.23)]; prolonged hospital stay for mothers (defined as > 3 inpatient hospital days for vaginal delivery or > 5 days for cesarean section), prolonged hospital stay for infants (defined as > 3 inpatient hospital days for vaginal delivery or > 5 days for cesarean section among children > 35 weeks gestational age), and method of delivery. Massachusetts Vital records collects maternal conditions from the hospital medical record and places them on the birth certificate. Infant and maternal demographic and diagnostic data obtained from birth certificates for this study included: age, race, education, parity, body mass index (BMI), year of birth, insurance, gestational diabetes, pre-gestational diabetes, preeclampsia/eclampsia, pregnancy induced hypertension, gestational age at delivery, birth weight, infant sex, preterm birth (gestational age < 37 weeks), and low birth weight (<2500 grams). We adjusted our model for all infertility diagnoses that were obtained from SART CORS: male factor; all female factors including endometriosis, PCOS, ovulation disorder, diminished ovarian reserve, uterine factor, tubal factor; unexplained infertility; and other reasons for ART and biopsy. This study was approved by the Institutional Review Boards at Dartmouth-Hitchcock and the Massachusetts Department of Public Health. The study was also approved by the SART Research Committee.

We report that embryo biopsy for preimplantation genetic diagnostic testing, in the context of current practice in the United States, does not increase the odds for adverse maternal and fetal outcomes related to placental dysfunction in singleton deliveries. Specifically, we found no increase in preeclampsia/eclampsia, pregnancy-related hypertension, preterm birth, low birth weight, placental abnormalities (placenta and vasa previa, placental abruption, placenta accreta/increta/percreta), gestational diabetes, or for cesarean section delivery when comparing biopsy to no biopsy.

Embryo biopsy for PGT followed by a single embryo transfer in a cryopreserved- thawed embryo cycle does not appear to increase outcomes related to placental dysfunction.

Preimplantation genetic testing (PGT) with assisted reproductive technology (ART) has increased recently in the United States, from 13% of ART cycles reporting PGT in 2014 to 37% of cycles reporting PGT in 2017 1 , 2 This procedure is commonly performed for older women undergoing IVF who are at greater risk for aneuploidy to lower miscarriage rate and increase delivery rate 3 The procedure is also performed to detect single gene defects and chromosomal rearrangements in reproductive age women of all ages 4 Contemporary practice of PGT in the United States generally involves culture of embryos to the blastocyst stage (day 5–7 after fertilization), biopsy of several trophectoderm cells that would become the placenta, transport of biopsy specimens off site for genetic testing, and cryopreservation of blastocysts until biopsy results are available 5 A single euploid embryo is then transferred into the uterus in a subsequent cryopreserved-thawed transfer cycle. Prior studies report mixed results with respect to embryo biopsy and maternal and neonatal outcomes related to placentation. Some studies report that PGT is associated with increased hypertensive disorders or small for gestational age infants 6 , 7 increased preeclampsia 8 , or increased preterm birth 6 , 9 while others report no such increases 10 , 11 , 12 . Results likely differ based on study of polar body, cleavage stage, or trophectoderm stage embryo biopsies, and the analysis of fresh combined with cryopreserved embryo transfers. Embryo cryopreservation is reported to increase preeclampsia diagnosis 13 , 14 and birth weight 15 compared to fresh embryo transfer. Our objective was to compare maternal and neonatal outcomes of singleton births following cryopreserved-thawed single embryo transfers, with and without embryo biopsy, to reflect the contemporary practice of PGT in the United States. Prior studies are mostly limited in size; our goal was to report a large study of both maternal and neonatal outcomes to assess safety of the procedure.