Objective: To compare pregnancy outcomes after single blastocyst embryo transfer among patients whose fi rst autologous embryo transfer was either a fresh embryo transfer or a frozen embryo transfer (FET) after a freeze-all, in the absence of preimplantation genetic testing for aneuploidy (PGT-A). Design: A multicenter retrospective cohort analysis. Setting: National multicenter fertility practice. Patient(s): A total of 8,319 autologous fi rst blastocyst embryo transfers in the absence of PGT-A were analyzed. Of them, 6,755 transfers were fresh embryo transfer (ET) and 1,564 transfers were FET after freeze-all. Exposure: Patients underwent either a fresh or a frozen initial autologous, single blastocyst transfer in the absence of PGT-A. Main Outcome Measure(s): Primary outcome measure was live birth rate. Secondary outcome measures included positive pregnancy test, clinical pregnancy rate, and miscarriage rate. Result(s): Live birth rate was comparable between the fresh ET and FET groups in the absence of PGT-A, after performing generalized estimating equation modeling to account for age, body mass index, antral follicle count, basal follicle-stimulating hormone, progesterone on day of trigger/day of fi nal lining check, peak estradiol during in vitro fertilization stimulation cycle, number of oocytes retrieved during ovarian stimulation cycle, and primary diagnosis. Similarly, the secondary outcome variables of positive pregnancy test, clinical pregnancy rate, and miscarriage rates were comparable between the cohorts. These fi ndings were observed in the entire study group, within each age category of < 35, 35-37, 38-40, and > 40 and among each stratified peak estradiol group. Conclusion(s): In the absence of PGT-A, patients and fertility providers can elect to pursue either fresh ET or embryo freeze-all with subsequent FET during the fi rst autologous in vitro fertilization cycle. (F S Rep (R) 2024;5:369-77. (c) 2024 by American Society for Reproductive Medicine.)

Aneuploidy is identified in at least 55% of products of conception of women with recurrent pregnancy loss (RPL) [1]. Preimplantation genetic testing for aneuploidy (PGT-A) has been suggested as a tool for treatment of patients with idiopathic RPL [2]. Data on the effectiveness of IVF with PGT-A in improving live birth rate (LBR) in women with idiopathic RPL have been contradictory [3,4,5], and it is unclear whether IVF with PGT-A is more effective than expectant management in achieving a live birth [2,6] in these patients. IVF with PGT-A is an expensive technology and may not show a cost-benefit until a patient has reached an advanced maternal age [7].

There is growing demand to apply IVF 'add-ons' hoping that it will improve Frozen Embryo Transfers (FBT) outcomes. Endometrial Receptivity Analysis (ERA) is one such test that attempts to predict optimal endometrial receptivity to personalize the timing of FBT. However, several recent data have called into question the efficacy of this test [1]. Recently, an RCT demonstrated that ERA does not improve live birth (LB) from single euploid FBT in an unselected population [2]. Research in this area is critical as ERA is costly, time intensive, and invasive.

Assisted hatching (AH) may optimize embryo implantation by opening the zona pellucida to enhance embryo attachment to the endometrium. AH has been employed for a variety of conditions that may predispose the embryo to impaired implantation, diminished oocyte quality leading to poor embryos, cryopreserved embryos, and repeat implantation failure. It has been suggested that AH be used in cycles where oocytes have been vitrified, due to possible physiologic changes in the zona induced by vitrification/warming1,2. However, studies to date have not been able to validate the benefit of this practice3. The purpose of this study was to evaluate the impact of AH on implantation and pregnancy rates among blastocyst transfers from vitrified donor eggs. Data from oocyte donor-recipient cycles was extracted from the Donor Egg Bank USA (DEBUSA) cycle report database from 29 separate centers. A standardized GnRH antagonist suppression protocol with a GnRH agonist trigger was used for ovarian stimulation. Donors met baseline criteria including age (21-32), BMI (18-28), and negative expanded carrier screening. Oocytes in metaphase II (M2) without visible defects were vitrified 38 hours post maturation trigger. Once selected from the DEBUSA database, oocytes were warmed and fertilized at their respective centers by IVF or ICSI. Assisted hatching was performed by one of the various described techniques in literature. Fresh embryo transfer was performed using standard technique at each respective center. Embryos from 1812 donor oocytes from January 1, 2016 through June 30, 2019 underwent AH (1334) or no AH (468). Each of the groups for this analysis were compared with Chi Square test to determine significance. Implantation rate is defined as the number of intra-uterine gestational sacs observed on the 5 to 7-week ultrasound and divided by the number of embryos transferred. The AH group had a 56.6% implantation rate. The non-AH group had a 66.1% implantation rate (p =0.0001). +HCG per transfer did not vary significantly between the AH vs non-AH groups (71.3% vs 75.6%, respectively, p= 0.0814). Clinical pregnancy rate (61.8% vs 69.6%) and ongoing pregnancy rate (51.8% vs 61.7%) were significantly lower while and SAB rate was significantly higher (15.9% vs 10.4 vs%) for transfers where AH was performed (p= 0.0021, 0.0002, and 0.0156, respectively). Table 1: Outcomes in AH vs No AH Vitrified Donor Oocyte Cycles This retrospective analysis suggests that AH may decrease odds of success from IVF using vitrified donor oocytes. IVF is an expensive process for patients, and additional procedures without clear benefit should be avoided. Additional prospective randomized trials are needed to enable firm conclusions regarding the use of AH for embryos from vitrified oocytes.

Recent RCT data indicate that programmed endometrial preparation regimens inclusive of intramuscular progesterone (IMP) are superior to regimens using vaginal progesterone (VP) only, in terms of live birth (LB) from frozen embryo transfer (FET) (1). Studies suggest inferior FET outcomes with lower serum progesterone concentration (P). A recent meta-analysis suggested P <10ng/mL as a negative prognosticator. However, most published analyses relied upon FET with VP only (2). It remains unclear whether P is associated with lower LB when IMP is used. To evaluate whether lower P was associated with lower LB following FET and whether the association differed depending upon the route(s) of progesterone used. 399, 388 and 210 transfers were randomized to the IMP, IMP+VP, and VP arms, respectively. P was assessed two weeks after FET, at the time of 'pregnancy test' (1). LB and secondary outcomes were assessed for patients with P < 10ng/mL ('low') vs. >/= 10ng/ml ('adequate') (2). Chi-square analyses were performed, and a value of <0.05 was considered statistically significant. P varied significantly based on route of administration. The VP arm had the lowest and the IMP arm had the highest mean P (1). 12%, 46.1% and 83.8% of patients had low P in the IMP, IMP+VP, and VP arms, respectively (Table 1). Miscarriage was higher, and clinical pregnancy and LB rates were lower among subjects with low P who received VP only (relative to those with low P receiving IMP or IMP+VP). While these differences were not statistically different when the between arm comparison was limited to those subjects with adequate P, power for this analysis was constrained by the small sample with adequate P following VP only (N=34). LB was only 30% among these subjects, versus 49% in the other IMP containing regimens (Table 2, upper). When evaluating outcomes for subjects with 'low' versus 'adequate' P within each treatment arm, there were no statistically significant differences in the VP or the IMP+VP arms. In the IMP only arm, those with low P had lower pregnancy (+hCG), clinical pregnancy, and live birth rates; however, the small sample with low P in the IMP arm (N=48) is limiting (Table 2, lower). RCT data demonstrated lower LB and higher miscarriage following programmed FET with VP only. VP resulted in significantly lower P. While low P was associated poorer outcomes in the overall cohort (2), this association was not clear when comparing low vs adequate P within treatment arms. Though limited by timing of serum P measurement (two weeks post FET rather than at time of FET) and low subject numbers with low P in the IMP arm and with adequate P in the VP arm, these secondary analyses suggest that poorer outcomes with VP only were unlikely attributable to inadequate serum P alone. Additional prospective studies are needed to elucidate the importance of P for FET, particularly for regimens inclusive of IMP.

To compare outcomes of natural cycle (NC) IUI versus OI/IUI among non-infertile women. Retrospective chart review. More than 77,000 IUI cycles performed at a single infertility center from 2004-2017 were reviewed. Single women and lesbian couples were included, as well as couples with male factor and no female factors if they used donor sperm or if the male partner produced a good sample for insemination. Women over 42 years or with subfertility diagnoses (including but not limited to endometriosis, PCOS, ovulatory dysfunction, tubal factor, uterine factor, and unexplained infertility) were excluded, as were inseminations with fewer than 8 million total motile sperm. Ovulation induction was by oral administration of clomiphene citrate or letrozole (no gonadotropins). Clinical pregnancy was defined by ultrasound confirmation of an intrauterine gestational sac, with multiple pregnancies having multiple gestational sacs. 8,315 IUI cycles were available for analysis (4,358 natural cycle and 3,957 OI). There were 1,312 clinical pregnancies (16%), including 80 twins and 5 triplets. While the pregnancy rate was significantly higher for OI versus natural cycle IUI in the 38-40 age group (14% vs 10%), pregnancy did not differ between OI and natural cycle IUI in the three other age groups, suggesting that the significance in the 38-40 year group may have occurred by chance. There were 13 twin pregnancies within the natural cycle IUI group (2% of all pregnancies) and 72 multiple pregnancies (11%), 67 of which were twins (10.1% of pregnancies) and 5 of which were triplets (0.8% of all pregnancies), in the OI/IUI group (p < 0.0001, Fisher's exact).Tabled 1Clinical Pregnancy Rate by Age in Natural Cycle IUI versus Ovulation Induction IUIAge (years)Natural CyclesClinical Pregnancy from NCNC Pregnancy per CycleOvulation Induction CyclesClinical Pregnancy from OIOI Pregnancy per CycleP-value comparing Clinical Pregnancies from NC vs. OI IUI (chi square)<35234640217.1%212040018.9%0.1335-37103316415.9%91114215.6%0.8638-40767739.5%72910314.1%0.00641-42212146.6%197147.1%0.84 Open table in a new tab There is no improvement in clinical pregnancy rates with OI versus natural cycle IUI among young, non-infertile women. However, clinical pregnancy rates were higher for older women (age 38-40) with OI/IUI versus natural cycle IUI. Multiple pregnancy rates are much higher with OI compared to natural cycle IUI. Our results indicate that natural cycle IUI may be just as efficacious as OI/IUI in the younger, non-infertile patient but with a reduced risk of multiples.

To compare IVF outcomes among African American versus Caucasian women. Retrospective cohort study. All retrievals for fresh autologous IVF from 2004-2016 at a private IVF center among women self-identifying as Caucasian or African American were reviewed. Unadjusted comparisons were evaluated by t-test or chi-square. Adjusted comparisons were performed with multivariate generalized estimating equations (GEE) analyses accounting for multiple retrievals per patient and adjusted for differences in age, body mass index (BMI), diagnoses, and treatment year. 36,217 retrievals for fresh autologous IVF were performed among Caucasian (29,547) and African American (6,670) women. African Americans were one year older, nearly 3kg/m2 higher BMI, more likely to have diminished ovarian reserve or uterine factor and less likely to have ovulation disorders (p <0.0001 for each). Total gonadotropin dosage did not differ between comparable Caucasians and African Americans. Adjusted GEE analyses indicated that African Americans had higher serum estradiol, thicker endometrium, more oocytes and more surplus good quality blastocysts (grade BB or better). Conversely, African Americans had lower pregnancy, higher clinical pregnancy loss, and lower live birth. By current standard clinical measures, African Americans respond better to IVF stimulation than comparable Caucasian women, with higher serum estradiol, thicker endometria, more oocytes, and more surplus high-quality blastocysts available after fresh embryo transfer. Despite these advantages, clinical pregnancy was 9% lower, clinical pregnancy loss was 24% higher, and live birth was 14% lower for African Americans relative to comparable Caucasians. Poorer IVF outcomes among African Americans may result either from aspects of embryo quality not captured by current grading systems or as yet unidentified uterine factors, suggesting these as potentially fruitful areas of future research.Tabled 1CaucasianAfrican AmericanP-valueCaucasian (adj)African American (adj)P-valueSerum estradiol at trigger (pg/mL)222.1 1110246.5 1275<0.0001220.9 1131257.4 1168<0.0001Endometrial thickness at trigger (mm)11.6 2.511.9 2.9<0.000111.6 2.611.9 2.7<0.0001Oocytes retrieved13.7 8.013.7 9.0NS13.5 8.014.5 8.3<0.0001Mature (MII) oocytes10.3 6.59.4 6.8<0.000110.7 6.710.7 6.8NSFertilized (2pn)7.9 5.47.2 5.5<0.00017.7 5.47.6 5.4NSEmbryos per transfer2.0 0.82.0 0.8NS2.0 0.72.0 0.8NSCryopreserved blastocysts1.2 2.31.4 2.6<0.00011.2 2.31.6 2.3<0.0001Positive hCG per transfer57.8%49.7%<0.000157.5%52.4%<0.0001Clinical pregnancy per transfer49.0%42.0%<0.000148.6%44.3%<0.0001Clinical pregnancy loss18.2%26.1%<0.000117.7%21.9%<0.0001Live birth per transfer40.0%31.0%<0.000138.8%33.3%<0.0001 Open table in a new tab

To compare singleton gestational age at birth between minority racial/ethnic groups (African American, Asian, and Hispanic) and Caucasian IVF patients. Retrospective cohort study All singleton IVF pregnancies ending in live birth among women self-identifying as Caucasian, African American, Asian, or Hispanic from 2004-2016 at a private IVF practice were reviewed. Gestational age at birth was calculated as the number of days from oocyte retrieval to birth, plus fourteen. Unadjusted comparisons between racial/ethnic groups were evaluated by t-test or chi-square. Generalized estimating equations (GEE) analyses accounted for parity and adjusted for differences in age, body mass index (BMI), infertility diagnoses, and treatment year. 10,371 singleton births were available for analysis. In unadjusted comparisons, African American births occurred over 6 days earlier compared to Caucasian births. Some of the shorter gestation among African Americans was explained by their higher BMI (p <0.0001) and higher incidence of uterine factor (p <0.0001), both of which are risk factors for earlier delivery. After adjusting for these and other demographic variables, African Americans still delivered 5.5 days earlier than Caucasians (p <0.0001). In adjusted GEE analyses, African American births were more than three times as likely to be either very preterm (2.9% vs 0.9%, p <0.0001) or extremely preterm (1.4% vs 0.4%, p <0.0001). Gestational ages of Asian and Hispanic births were both comparable to Caucasian births. Unlike other minority groups (Asians and Hispanics), ART singleton live births occurred at a significantly earlier gestational age in both unadjusted and adjusted comparisons of African American versus Caucasian births. The higher risk of preterm delivery in African Americans seen in unassisted conceptions also exists in ART conceptions and persists despite adjustment for known confounders.Tabled 1CaucasianAfrican AmericanAsianHispanicNumber of Births710512041601461Age at Birth, days, mean SD272.7 13.7266.4 19.7*271.9 13.0271.2 14.7Term ( 37 weeks)90.4%80.9%*90.0%86.1%Moderately Preterm (32-37 weeks)8.4%14.2%*9.1%12.2%Very Preterm (28-32 weeks)0.9%3.2%*0.6%1.3%Extremely Preterm (< 28 weeks)0.4%1.8%*0.4%0.4%* p < 0.0001 versus Caucasian Open table in a new tab

The aim of this study was to determine if multiple vitrifications or biopsies for preimplantation genetic testing for aneuploidy (PGT-A) have an effect on pregnancy outcomes. Retrospective cohort study Pregnancy outcomes from the transfer of blastocysts biopsied from 2016 through 2017 were analyzed. Transfers of blastocysts which were biopsied, vitrified, and thawed (Standard PGT-A) were compared to embryos which were initially vitrified then thawed, biopsied, and vitrified a second time in order to do PGT-A (Twice Vitrified & Single Biopsy). Embryos requiring a second biopsy due to inadequate DNA for amplification or nonconcurrent results (Twice Vitrified & Twice Biopsied) were also analyzed. Embryos created from donor eggs were excluded as well as transfers using gestational carriers. Outcomes included positive hCG, biochemical loss, clinical pregnancy loss, and live birth. Analysis was performed using unadjusted and adjusted GEE models accounting for number of prior IVF cycles and number of embryos transferred. 6,153 embryo biopsies were performed and 970 euploid embryo transfers were available for analysis. This included 863 with standard PGT-A, 86 twice vitrified and single biopsy, and 21 twice biopsied and twice vitrified. Compared to the standard PGT-A group, patients with twice vitrified & single biopsied embryos were younger and had fewer prior cycles. After adjusting for age and number of prior cycles, all three groups had similar initial pregnancy rates. However, there was increased biochemical loss and spontaneous abortion in the twice vitrified and single biopsy group. As a result of loss, there was an absolute reduction in live birth by 15.6% in embryos in the twice vitrified and single biopsy group. Live birth rate had an absolute reduction by 21.9% in the twice vitrified and twice biopsied group, although this did not achieve statistical significance due to a lack of power. In this study, euploid embryos vitrified twice resulted in fewer live births than those vitrified once prior to transfer. We must acknowledge, that patients seeking PGT-A post hoc (the twice vitrified cohort) may represent a poorer prognosis group. To our knowledge this is the largest cohort to analyze pregnancy outcomes in embryos undergoing multiple vitrifications and biopsies. Further research is needed to determine if this difference is a direct effect of multiple vitrifications or patient characteristics.

Ovarian stimulation using a gonadotropin-releasing hormone (GnRH) antagonist protocol offers the flexibility of utilizing a GnRH agonist (GnRHa) or hCG trigger for final oocyte maturation. When the follicle number or estradiol (E2) concentration is low, an hCG trigger may be selected if it is determined that the patient is at low risk of ovarian hyperstimulation syndrome (OHSS). It is unclear whether a GnRHa or hCG trigger in this clinical scenario yields superior outcomes. To compare embryologic and clinical outcomes between GnRHa and hCG triggers in a patient population at low risk of OHSS. This is a retrospective cohort study of patients undergoing autologous in vitro fertilization (IVF) at a large, private fertility center. All patients underwent ovarian stimulation using a GnRH antagonist protocol. The choice of trigger medication was at the discretion of the primary physician. Patients were divided into three groups based on the number of follicles >12mm on the day of trigger: ≤ 5 (Group 1), 6-10 (Group 2), and 11-15 (Group 3) follicles. Luteal support consisted of vaginal progesterone and oral estradiol. Patients who received a GnRHa trigger were also given 1,500 IU hCG immediately following egg retrieval. Outcome variables included: total number of oocytes retrieved, number of metaphase II (MII) oocytes, %MII oocytes, number of good quality blastocysts (AA or AB blastocysts by Gardner and Schoolcraft grading system [1]), and live birth rate per cycle. GEE analyses accounted for multiple cycles within a given patient and adjusted for patient age, serum E2 on day of trigger, and the number of follicles >12mm on day of trigger. 8,001 patients were included in the analysis. Outcomes for the three groups are presented in Table 1. There were significantly more oocytes retrieved and good quality blastocysts after GnRHa vs. hCG trigger in all three groups, even after adjusting for follicle number and E2. The %MIIs were similar between the two trigger types in all groups. The number of MII oocytes and the live birth rate were significantly higher after GnRHa vs. hCG trigger in Groups 2 and 3.Tabled 1Follicles >12mm on Day of TriggerGnRHa Trigger(n=2,468)hCG Trigger(n=5,533)P value≤ 5n=13 (1.5%)n=865 (98.5%) Oocytes retrieved (n)8.5 (5.2-11.8)4.3 (4.2-4.4)0.01MII (n)5.4 (2.7-8.1)3.3 (3.2-3.4)0.12% MII65% (52-77%)76% (75-78%)0.07 Quality Blastocysts (n)1.8 (0.4-3.2)0.3 (0.3-0.4)0.03 Live birth rate (%)25% (10-52%)13% (11-16%)0.206-10n=723 (19.7%)n=2,955 (80.3%) Oocytes retrieved (n)12.3 (11.8-12.7)10 (9.8-10.2)<0.001MII (n)9.5 (9.1-9.9)7.7 (7.5-7.8)<0.001% MII79% (77-81%)78% (77-78%)0.11 Quality Blastocysts (n)2.1 (1.9-2.3)1.4 (1.3-1.4)<0.001 Live birth rate (%)28% (25-32%)24% (23-26%)0.0411-15n=1,732 (50.3%)n=1,713 (49.7%) Oocytes retrieved (n)16.4 (16.1-16.8)15.7 (15.4-16)0.001MII (n)12.8 (12.5-13.1)12.1 (11.9-12.4)0.001% MII79% (78-79%)78% (77-79%)0.36 Quality Blastocysts (n)2.9 (2.7-3.0)2.6 (2.4-2.7)0.002 Live birth rate (%)30% (28-33%)26% (24-29%)0.02*Mean (95% confidence interval). Adjusted for age, E2, and follicles >12 mm. Open table in a new tab *Mean (95% confidence interval). Adjusted for age, E2, and follicles >12 mm. This analysis of a large number of GnRH antagonist IVF cycles suggests that patients at low risk of OHSS have no impairment in outcomes after GnRHa trigger compared to hCG trigger. These data suggest that some embryologic outcomes and live birth may be improved after GnRHa trigger, though the absolute differences were small. It is possible that a GnRHa trigger may benefit oocyte quality via an endogenous LH and FSH surge. However, there were no differences in % mature oocytes, and the limitations of the retrospective study design do not allow for definitive conclusions. This can only be adequately addressed with randomized controlled trials. These data do not suggest a detriment in using GnRHa trigger to patients at low risk of OHSS.

Estimate the risk of multiple gestation in OI-IUI with increasing mature follicle number based on patient age. Retrospective cohort. 50,492 IUI cycles from 2004 through 2017 at a large private practice were analyzed. The number of follicles ≥ 14 mm were measured on the day of trigger. All cycles were included regardless of stimulation protocol or diagnosis. Cycles with a total motile sperm count less than 8 million were excluded. Clinical pregnancy rates per IUI cycle were recorded. Singleton pregnancy per IUI cycle, twin pregnancy per clinical pregnancy (CP), and triplet pregnancy per CP were categorized by SART patient age groups. Clinical pregnancy rates ranged widely from 3.7% to 24.4% per cycle and decreased with age. In women under 35, increasing the number of follicles from 1 to 5 increased clinical pregnancy by 7.8%, but increased per pregnancy twin and triplet risk by 4 fold and nearly 30 fold, respectively (P<0.01 each). In this age group, patients with 3 or more follicles, 21% of all pregnancies were multiple gestation and 6% of all pregnancies were triplets. In women up to 40, increasing follicle numbers from 1 to 2 increased clinical pregnancy by only 3% but with the expense of a nearly 3 fold risk in twins and triplets per pregnancy (P<0.01 each). In women over the age of 40, up to 3 follicles nearly doubled the likelihood of pregnancy while maintaining an 11% risk of multiple gestation per pregnancy.Table 1CP rates per IUI resulting in singleton, and CP per pregnancy for multiples< 35 years, n=26,52235-37 years, n=11,31738-40 years, n=8,17341-42 years, n=3,05443-44 years, n=1,426Pregnancy TypeCP per IUI/Singleton per IUI/Twin per CP/Triplet+ per CPCP per IUI/Singleton per IUI/Twin per CP/Triplet+ per CPCP per IUI/Singleton per IUI/Twin per CP/Triplet per CPCP per IUI/Singleton per IUI/Twin per CP/Triplet per CPCP per IUI/Singleton per IUI/Twin per CP/Triplet per CP1 follicle, n=16,84416.6%/ 15.8%/4.4%/0.3%14.3%/13.5%/5.5%/0%10.5%/9.9%/4.8%/0.4%5.7%/5.5%/3.9%/0%4.1%/3.7%/10.0%/0%2 follicles, n=16,60220.1%/17.3%/13.2%/1.0%18.2%/15.7%/13.0%/0.6%13.9%/11.8%/13.9%/1.1%9.4%/8.3%/10.2%/1.1%4.2%/3.7%/10.5%/0%3 follicles, n=10,53921.0%/16.0%/20.3%/3.7%20.1%/16.7%/15.2%/1.8%16.9%/13.9%/15.3%/2.5%11.1%/10.2%/6.8%/1.4%8.7%/8.0%/4.2%/4.2%4 follicles, n=4,80922.1%/15.7%/23.6%/5.7%21.1%/15.5%/24.5%/2.1%19.5%/15.8%/15.1%/3.8%15.1%/11.7%/22.4%/0%8.3%/6.8%/18.2%/0%5 follicles, n=1,69824.4%/17.7%/18.7%/8.8%21.9%/13.3%/27.9%/11.6%19.7%/13.9%/25.4%/4.2%19.6%/16.1%/15.2%/3.0%10.7%/8.0%/25.0%/0% Open table in a new tab Caution should be used in ovulation induction with more than 2 mature follicles in women under the age of 40, as more follicles does not improve singleton pregnancy rate and substantially increases multiple gestation. Women over the age of 40 have improved pregnancy rates with multiple follicles, and the multiple gestation rate does not substantially rise until 4 or more follicles are present. This large dataset can be used to guide patient counseling and clinical decision making.

To compare three different progesterone (P) replacement protocols for vitrified-warmed blastocyst transfer (VBT). Three-arm non-inferiority randomized controlled trial (RCT). Planned transfers of vitrified-warmed high quality (Grade BB or better at time of vitrification) un-biopsied blastocyst(s) were randomized to receive: (1) 50 mg/d IM P only; (2) 200 mg twice daily vaginal Endometrin; or (3) 200 mg twice daily Endometrin plus 50 mg IM P every third day. The primary outcome was live birth per transfer. Secondary outcomes included pregnancy (positive hCG 2 weeks after transfer), clinical pregnancy (ultrasound confirmation of intrauterine gestational sac 4-5 weeks after transfer), biochemical pregnancy losses, and clinical pregnancy losses. Outcomes were compared among treatment arms by three-group X2, followed by pairwise X2 comparisons as appropriate. 997 cycles were randomized, underwent VBT, and administered the study medications per assigned protocol. A planned interim analysis, performed once 50% of subjects had completed the final study visit, revealed significantly lower ongoing pregnancy in the Endometrin only arm. This arm only was unblinded and enrollment terminated. The study was completed according to our predetermined enrollment goals in the two remaining arms, which remained blinded to clinicians and the statistician until after analysis was complete. There were no differences in any treatment outcomes between the daily IM P only arm and the daily Endometrin plus IM P every third day arm. There was a 40% reduction in live birth in the Endometrin only arm relative to cycles that included IM administration of P (28.6% vs 47.3%). This difference in live birth was primarily due to a biochemical loss rate that was more than twice as high for Endometrin only compared to the groups administering IM P (33% vs 15%). Increased clinical pregnancy loss and, to a lesser extent, decreased implantation (postitive HCG) likely also contributed to the poorer birth outcomes with Endometrin only. Live birth following VBT with vaginal-only P replacement was reduced by 40% relative to VBT where IM P was used. This difference was attributable to an increased rate of pregnancy loss. These RCT data indicate that vaginal-only P replacement for VBT should be avoided. In contrast, pregnancy and birth outcomes are equivalent between daily IM P or daily vaginal P supplemented with IM P every third day.Tabled 1Outcomes by Treatment Arm among Cycles Completed Per ProtocolIM P onlyEndometrin + IMPEndometrin onlyOverall Chi-square (p val)IMP only vs. Endometrin only (p val)IMP only vs Endometrin + IMP (p val)IM + Endometrin vs. Endometrin only (p val)Vitrified Blastocyst Transfers (N)399388210Positive hCG per transfer (%)68.964.959.00.050.0150.240.15Biochemical loss per positive hCG (%)17.113.133.1<0.0001<0.00010.20<0.0001Clinical pregnancy per transfer (%)57.156.439.5<0.0001<0.00010.84<0.0001Pregnancy loss per clinical pregnancy (%)14.918.727.70.0360.010.280.09Total pregnancy loss (biochemical plus clinical) per positive hCG (%)29.529.451.6<0.0001<0.00010.98<0.0001Live birth per transfer (%)48.645.928.6<0.0001<0.00010.44<0.0001 Open table in a new tab

To evaluate outcomes of vitrified euploid blastocyst transfers according to the number of previous consecutive unsuccessful cryopreserved euploid blastocyst transfer cycles Retrospective cohort study All patients undergoing autologous transfer of cryopreserved blastocysts diagnosed as euploid by preimplantation genetic testing for aneuploidy (PGT-A) from 2012-2016 at a large assisted reproduction center. Clinical pregnancy was defined by ultrasound confirmation of an intrauterine gestational sac. A total of 1489 patients underwent one or more transfers of PGT-A euploid cryopreserved blastocysts during the study period. 387 of these patients underwent 1 to 5 additional attempts after initial failures. While age and BMI did not differ by attempt number, the number of embryos per transfer increased with increasing numbers of previous failures. The likelihood of pregnancy outcomes declined progressively with additional attempts following initial failures. Live birth declined from 52% in first time euploid transfers to only 20% in the fifth transfer cycle after failed euploid blastocysts transfers. Compared to first transfers of euploid blastocysts, the relative reductions in birth rate per transfer was 22%, decreasing from 52% to 41%, for second attempts after one failed euploid blastocyst transfer. After two prior unsuccessful transfers of euploid blastocysts, birth per transfer decreased by an additional 33%, from 41% to 35%. Moreover, this decline in success was observed despite transferring a greater number of embryos in subsequent cycles. The cumulative result was that the number of live born children per transferred euploid blastocyst declined from 49% to 18% from the first to the fifth attempts. Birth rates from euploid blastocyst transfers declined rapidly with an increasing number of previous unsuccessful transfers of euploid blastocysts. Patients who fail to achieve live birth after multiple euploid blastocyst transfers have a poor prognosis and warrant further clinical investigation.Tabled 1Euploid Blastocyst Transfer1st2nd3rd4th5th6thP-valueTransfers14893871022651Age (years)36.236.036.535.535.234.00.78BMI (kg/m2)25.525.425.727.425.023.50.54Embryos per transfer1.11.21.31.52.21.0<0.0001Positive hCG72%69%58%65%80%100%0.024Biochemical Losses12%18%17%18%50%100%0.002ClinicalIntrauterine Pregnancy63%57%48%54%40%0%0.0007ClinicalPregnancy Losses17%28%28%36%50%<0.0001Live Birth52%41%35%35%20%0%<0.0001Children per embryo transfer49%36%28%26%18%0%<0.0001 Open table in a new tab

Theoretically oocyte cryopreservation has the potential to disrupt the meiotic spindle leading to abnormal segregation of chromosomes and increased aneuploidy in embryos. Therefore, PGS might have benefit in vitrified donor egg cycles. In contrast, embryos derived from young donor oocytes are expected to be predominantly euploid, and trophectoderm biopsy may have a negative effect relative to transfer without biopsy. We evaluated the rate of aneuploidy among blastocysts derived from vitrified donor oocytes and assessed clinical outcomes resulting from transfer of PGS normal blastocysts vs. from transfer of PGS untested embryos. Retrospective cohort study. 2012 -2017 vitrified donor egg cycles performed at multiple clinics were reviewed. Outcomes from the same donor's oocytes that were used in both PGS (study group) vs. non-PGS (control group) recipient cycles were compared. Importantly, this analysis enabled a single donor to serve as her own control. GEE models adjusted for multiple cycles and controlled for PGS and non-PGS cycles from the same donor. PGS was performed by array comparative genomic hybridization or DNA microarray. 515 recipient cycles from 96 donors were available for analysis. 32% of donor oocyte embryos were aneuploid. 11.1% of PGS cycles had no euploid embryos for transfer. These cycles were included in all analyses. Among all cycles, clinical and total pregnancy loss rates were higher in the non PGS group compared to the PGS group. However, there was no statistically significant improvement in the endpoint of ongoing/live birth with PGS testing. Cumulative pregnancy rates were 53% without PGS and 53% with PGS (Table: Comparisons of vitrified donor egg IVF with no PGS vs. with PGS using GEE models for estimation of the means (a) Analysis limited to the first transfer resulting from insemination of a lot of vitrified donor oocytes, (b) Analysis of the initial and all subsequent transfers (C) Analysis of cumulative pregnancy). Aneuploidy rate is relatively low in donor oocyte cycles. PGS does not improve ongoing pregnancy or birth rates in the recipients of vitrified donor oocytes. This finding was persistent in analyses of the first transfer, all transfers, and cumulative pregnancy rates. While PGS does decrease the total pregnancy loss rate in donor oocyte recipient patients, it is at the expense of additional cost and time of PGS testing in this population.Tabled 1OutcomeNo PGSPGSOR (95% CI)P valueFIRST TRANSFER (N)319107Ongoing or Birth44.346.91.10 (0.69-1.78)0.66Clinical pregnancy loss11.34.30.34 (0.08-1.53)0.16Biochemical loss13.310.80.79 (0.28-2.16)0.64Total Pregnancy Loss24.515.10.56 (0.27-1.33)0.19ALL TRANSFERS (N)395120Ongoing or Birth43.547.91.19 (0.76-1.85)0.44Clinical pregnancy loss12.53.70.27 (0.06-1.19)0.08Biochemical loss13.48.40.71 (0.26-1.93)0.5Total pregnancy loss25.912.10.41 (0.17-0.99)0.04CUMULATIVE Ongoing or Birth53.052.80.99 (0.62-1.58)0.97 Open table in a new tab

The percentage of reproductive age women classified as obese has increased over the past decades. Obesity has been associated with several adverse pregnancy outcomes including lower implantation rates and pregnancy loss. Various theories have been proposed to identify a cause including poor embryo quality and endometrial receptivity. We attempted to assess an association between maternal BMI and live birth in frozen embryo transfers with known euploid embryos. Retrospective cohort. Autologous cycles with embryos undergoing preimplantation genetic screening (PGS) between March 2014 and July 2015 were available for analysis. Live birth outcomes from the first FET of a PGS-normal blastocyst(s) after retrieval were included in this analysis. PGD for single gene disorders or known parental translocations were excluded from the analysis. The association of BMI and live birth were assessed using GEE models, accounting for multiple cycles within the same patients and adjusting for patient age and the number of embryos transferred. Outcomes were also analyzed using adjusted GEE models according to BMI categories per WHO guidelines; underweight (<18.5), normal (18.5-24.9), overweight (25-29.9), and obese (>30). 376 FET cycles from 361 unique patients were available for analysis. The average age in this patient population was 36.1 years. The number of embryos transferred and clinical pregnancy rates did not differ significantly in any BMI category. There was no association between BMI and live birth when BMI was analyzed as a continuous variable (OR 1.02, 95% CI 0.98-1.06, P=0.31). Patients with normal BMI had no significant difference in live birth rate when compared to overweight (OR 0.81, 95% CI 0.47-1.40, P=0.81) and obese patients (OR 1.18, 95% CI 0.67-2.07, P=0.55). BMI was not associated with clinical pregnancy or live birth in patients undergoing euploid FET in this cohort when analyzed as both a continuous and dichotomous variable. These results suggest that BMI may not have a significant impact in FET cycle outcomes when known euploid embryos are transferred.Tabled 1Underweight (n=8)Normal (n=206)Overweight (n=87)Obese (n=75)Age (years)37.735.236.636.1Average Embryos Transferred11.21.21.2Clinical Pregnancy Rate56%56%53%60%Live Birth Rate44%46%41%51% Open table in a new tab

Evidence suggests embryos can signal the endometrium and embryo quality may affect endometrial receptivity. Our objective was to determine if transferring a poor quality embryo with a high grade blastocyst is detrimental relative to transferring a high grade blastocyst alone (SET). Retrospective cohort study. Autologous IVF cycles from 2013-2015 were included. Patients with a double embryo transfer (DET) of a good quality blastocyst plus a fair or poor blastocyst, early blastocyst, or morula were compared to SET. The primary outcome was live birth. GEE models were used to account for multiple patient cycles and age. The primary analysis was in patients with 1 good blastocyst to transfer, additional lower quality embryo(s) available, and no suitable blastocysts to vitrify. Additional analyses were performed : (1) not limiting the cohort to those with only 1 good blastocyst (with GEE linear adjustment for supernumerary vitrified embryos as this was associated with birth) (2) by the stage of the second embryo (3) in women <38 (4) in women ≥38. 4,640 IVF cycles were analyzed. Patients with a DET were 2.5 years older than SET (P<0.01). In none of the analyses did transferring a second poor quality embryo negatively affect birth rate. In the primary analysis, transferring a second poor embryo increased live birth by 6% and the twin rate by 15%. Analyzing by stage, the addition of a poor quality blastocyst or early blastocyst markedly increased the twin rate by 22-27% with slight increases in live birth. The addition of a morula did not increase live birth but resulted in 12% more twins. In women <38, DET increased birth by 7% but resulted in 18% more twins. In women ≥ 38, DET increased pregnancy by 9% with a 15% increase in twins.Tabled 1nClinical PregnancyLive BirthTwinsAge adjusted birth OR (95% CI)Age adjusted birth P valuePrimary analysis: 1 good blastocyst available SET vs DETSET25550%44%1%RefRefDET28854%50%16%1.28 (0.87-1.90)0.20Subanalysis 1: SET vs DETSET375157%49%1%RefRefDET88961%56%21%1.34 (1.10-1.53)0.002Subanalysis 2: SET vs DET by embryo stageSET375157%49%1%RefRefDET w/Poor Blastocyst22967%61%27%1.59 (1.189, 2.14)0.002DET w/ Early Blastocyst45563%57%22%1.37 (1.08-1.64)0.006DET w/ Morula20556%50%12%1.02 (0.77-1.34)0.89Subanalysis 3: SET vs DET in women <38 years old<38 SET346258%51%1%RefRef<38 DET52963%58%19%1.32 (1.09-1.60)0.004Subanalysis 3: SET vs DET in women ≥38 years old≥38 SET28946%33%0%RefRef≥38 DET36055%45%15%1.38 (1.00- 1.91)0.004 Open table in a new tab Addition of a poor quality embryo did not have a detrimental effect relative to transfer of a single high quality blastocyst. In patients <38 whom ASRM recommends SET, transferring a second poor quality embryo increases the risk of twins with minimal increase in live birth.

While ample prospective data have shown vaginal progesterone to be adequate luteal and early pregnancy support for 'fresh' embryo transfer1, it remains unknown whether it provides adequate progesterone replacement for FBT2. Therefore, we set out to assess the non-inferiority of vaginal progesterone to daily intramuscular progesterone for replacement in programmed FBT in terms of live birth. In addition we set out to assess the non-inferiority of vaginal progesterone in combination with intramuscular progesterone every third day. In order to enable early recognition of inferiority if present, an a priori interim analysis was planned and completed once 50% of the total enrollment goal had completed the study. The interim analysis presented here assessed OPR as the primary outcome. Three-arm non-inferiority RCT. Subjects planning transfer of vitrified-warmed high quality (Grade BB or better at time of vitrification) un-biopsied blastocyst(s) were randomized to receive: (1) 50 mg/d intramuscular progesterone only; (2) 200mg twice daily vaginal Endometrin; or (3) 200 mg twice daily Endometrin plus 50 mg intramuscular progesterone every third day. At the time of planned interim analysis, OPR was assessed, along with pregnancy, clinical pregnancy, biochemical pregnancy loss, and clinical pregnancy loss. The study team including the statistician were blinded to the identity of the treatment arms, which were randomly labeled 'A,' 'B,' and 'C' in the dataset. 645 cycles were randomly assigned to one of the three treatment arms, underwent at least one dose of progesterone according to this assignment and went on to FBT. These were included in the intention-to-treat analysis (Table). OPR was more than one-third lower in Group C than in the other two groups (P<0.0001). This difference was attributable primarily to higher pregnancy loss. A per-protocol analysis was also conducted, and results were nearly identical. Unblinding to the identity of group C only revealed it to be the vaginal progesterone only group.Tabled 1Intention to treat; *Arm C was revealed to be vaginal P alone once results showed inferiorityABC*Overall χ2A vs. BA vs. CB vs. C# of transfers218217210+hCG141 (65%)143 (66%)126 (60%)p=0.41Biochemical losses18 (13%)28 (20%)42 (33%)p=0.0002p=0.12p<0.0001p=0.010Clinical pregnancy123 (56%)115 (53%)84 (40%)p=0.002p=0.47p=0.001p=0.007Clinical pregnancy losses13 (11%)13 (11%)19 (23%)p=0.038p=0.86p=0.019p=0.032Overall pregnancy losses31 (22%)41 (29%)61 (48%)p<0.0001p=0.20p<0.0001p=0.001Ongoing pregnancy110 (50%)102 (47%)65 (31%)p<0.0001p=0.47p<0.0001p=0.001 Open table in a new tab Relative to regimens inclusive of intramuscular progesterone, vaginal-only progesterone replacement for FBT results in lower OPR and should be avoided. The difference was attributable to an increased rate of pregnancy loss. Randomization to the vaginal-only arm was terminated with these findings. This RCT is ongoing to assess the non-inferiority of the vaginal plus every third day intramuscular progesterone arm vs. daily intramuscular progesterone in terms of live birth.

Several studies have demonstrated that day 5 blastocysts have higher implantation rates than day 6 blastocysts in fresh embryo transfers. It is unclear if this is due to poor embryo quality with delayed blastulation, or dyssynchronous endometrium. We aimed to assess embryo quality through preimplantation genetic screening (PGS) and determine if day 6 blastocysts had higher rates of aneuploidy. Retrospective cohort. Fresh autologous IVF cycles from March 2014 through July 2015 with embryos undergoing PGS were available for analysis. Patients who elected for PGD for single gene disorders or known parental translocations were excluded from the analysis. Mixed-effects logistic regression models were used to assess for an association between day of embryo biopsy and aneuploidy. Analysis was performed adjusting for maternal age using piece-wise model. Unadjusted analysis was also performed. Available for analysis were 707 cycles from 544 unique patients. 3,055 embryos were biopsied and stratified according to blastulation rate. An average of 4.3 embryos were biopsied per completed cycle, with similar numbers blastulating and biopsied on days 5 and 6. The average age in this population was 37 years. Prior to adjusting for maternal age, embryos biopsied on day 6 were more likely to be aneuploid than those reaching the blastocyst stage on day 5 (OR 1.246 (1.064-1.460), p =0.006). Aneuploidy was significantly associated with maternal age (≤37 years old OR 1.05 (1.02-1.09) p=0.002; >37 years old OR 1.26 (1.13-1.41) p<0.001). After adjusting for maternal age, embryos biopsied on day 6 continued to have higher aneuploidy rates (OR 1.169 (1.003-1.363), p=0.046). Our data demonstrated embryos with delayed blastulation are more likely to be aneuploid. To our knowledge this is the largest cohort with adequate power to address this question. Though this difference is statistically significant, the absolute difference is small, therefore factors other than aneuploidy may also contribute to lower pregnancy rates in day 6 fresh transfers.Tabled 1Day 5 Blastocyst (n=1538)Day 6 Blastocyst (n=1517)No. Biopsied/Cycle2.22.2Unadjusted Aneuploidy Rate53%(50%-56%)58%(55%-61%)Adjusted Aneuploidy Rate56%(53%-59%)60%(57%-62%) Open table in a new tab

Continued improvement in vitrification has made frozen embryo transfer (FET) a viable alternative to fresh embryo transfer, and use of a 'freeze all' strategy for ART has increased. Supraphysiologic steroid hormones resulting from ovarian stimulation for fresh IVF may adversely affect implantation. Preparation for FET may more closely mimic the 'natural' hormonal milieu, leading to improved cycle outcomes. Our objective was to use SART data to compare live birth rates from FET vs. those from fresh embryo transfer for both autologous and donor IVF cycles and to evaluate trends over time. Retrospective cohort study. All 2003 to 2014 SART cycles were reviewed. For autologous cycles, patients 35 years and younger were included. FET cycles for autologous and donor cycles included all transfers of frozen embryos resulting from the insemination of fresh or frozen eggs. Fresh donor cycles were those in which neither eggs nor embryos were ever frozen. Fresh autologous cycles included embryos derived from fresh and frozen oocytes. A total of 160,140 autologous and donor cycles were included. For autologous cycles, the difference in live birth consistently favored fresh transfer; however, the difference narrowed markedly over the 12 year period analyzed, from 14% in 2004 to only 6% 2014. Conversely, live birth rates from fresh transfer of embryos from donor eggs were 15-20% higher than those from frozen embryo transfer of embryos from donor eggs over the same duration (Table). This analysis of SART data demonstrates that the success of autologous FET now approaches that of fresh embryo transfer. This is likely due to improvement in cryopreservation methods as well as increased use of PGS, particularly among FET cycles. In contrast, fresh donor egg transfers demonstrated consistent and considerably higher live birth rates than donor egg FET. It is unclear why embryo cryopreservation is associated with a greater detriment when using embryos from donor eggs. This national phenomenon of poorer outcomes from FET following donor egg IVF warrants attention and further exploration.Tabled 1Percent of transfers resulting in live birth in FET and fresh donor/autologous IVF cycles.YearDonor FETDonor fresh transferDifferenceAutologous FETAutologous fresh transferDifference200430.050.920.929.543.313.8200530.550.419.930.542.512.0200632.152.1.20.033.144.911.8200732.154.122.034.046.112.1200832.755.022.335.647.311.7200933.855.121.335.647.511.9201034.855.620.838.747.89.1201135.754.919.239.346.37.0201237.256.619.442.447.14.7201340.556.115.644.447.73.3201438.453.615.242.749.36.6 Open table in a new tab