Oocyte utilization rate as pregnancy potential indicator (a multicentric study)

DESIGN: A retrospective cohort study. MATERIALS AND METHODS: All non-polycystic ovary patients who underwent their first frozen-thawed embryo transfers in our unit and had basal serum AMH evaluated between 2010 and 2015 were evaluated in this retrospective study. AMH was measured using the enzyme amplified two-site immunoassay (ELISA) provided by Beckman Coulter (AMH Gen II ELISA, Beckman Coulter). Individualized controlled ovarian hyper-stimulation (COH) protocols included long GnRH agonist, GnRH antagonist and minimal stimulation protocols, based on individual patient characteristics. When at least two follicles reached 18 mm in diameter, 5000-10,000 IU hCG (Ovidrel, Merck Serono) was adopted to achieve final follicular maturation and oocyte retrieval performed 32-36 hours later. Intracytoplasmic sperm injection was performed if the concentration of motile sperm was <1
106/mL, otherwise a conventional in vitro fertilization was used. Vitrification was performed for embryo freezing with the use of a Cryotop (Kitazato Corp.) device. Embryo transfer was operated under trans-vaginal sonography guidance. The primary outcome is Live birth rate, which was defined as rate of deliveries that resulted in at least one live born baby per transfer cycle. Chi-square and binary regression were used for data analysis. A P<0.05 indicated significant result. RESULTS: In total, 828 patients were included. We grouped them into three groups based on their baseline AMH concentration: low AMH group (<25th), middle AMH (25th- 75th) and high AMH level (>75th). The results showed lower AMH level was associated with significantly lower implantation rate (21.9% vs. 43.2% vs. 58.8%, P<0.001), clinical pregnancy rate (32.0% vs. 55.2% vs. 65.7%, P<0.001), live birth rate (21.8% vs. 43.6% vs. 52.7%, P<0.001) but higher miscarriage rate (31.8% vs. 17.5% vs. 15.4% P¼0.014). After stratifying the patients into two age groups, being younger or older than 35 years, we still observed significant influence of AMH on outcomes in all patients in both age groups. Furthermore, serum AMH kept significant in multivariate analysis when adjusting covariates (i.e. age, FSH, AFC, endometrium thickness, endometrium preparation protocols, number of embryos transferred, etiology of infertility). The area under the curve (AUC) for serum AMH, age, AFC and FSH were 0.635, 0.634, 0.615 and 0.543 respectively, for predicting live birth. CONCLUSIONS: Our results demonstrated that baseline AMH was an independent predictive factor of live birth rate of frozen embryo transfers, irrespective of maternal age. It had only moderate predictive value on predicting live birth, superior to that of AFC and FSH. P-186 Tuesday, October 18, 2016 ADVERSE PERINATAL OUTCOMES ASSOCIATED WITH FERTILITY TREATMENT IN LATE PRETERM INFANTS. L. W. Sundheimer,a E. T. Wang,b C. Quant,c C. Spades,d e fa C. F. Simmons, M. D. Pisarska. OB/GYN, Division of Reproductive Endocrinology and Infertility, Cedars-Sinai Medical Center, Los Angeles, CA; b Obstetrics and Gynecology, Cedars Sinai Medical Center, Los Angeles, CA; cFamily Medicine, Cedars-Sinai Medical Center, Los Angeles, CA; d OB/GYN, Cedars-Sinai Medical Center, Los Angeles, CA; ePediatrics, Cedars Sinai Medical Center, Los Angeles, CA; fOB/GYN, Division REI, Cedars-Sinai Medical Center, Los Angeles, CA. OBJECTIVE: To assess whether late preterm infants conceived by in vitro fertilization (IVF) and non-IVF treatment (NIFT) are at risk for adverse perinatal outcomes compared to spontaneously conceived infants. DESIGN: Retrospective cohort study. MATERIALS AND METHODS: We included all singletons delivering between 34 0/7 to 36 6/7 gestational weeks at our center from January 1, 2013 to December 31, 2014. Mode of conception (spontaneous, IVF or NIFT) was determined from an electronic chart review of delivery, perinatal, and neonatal intensive care unit (NICU) reports. Standard descriptive statistics including ANOVA, Kruskal-Wallis, and Chi-square tests were conducted to compare the three groups. The primary outcome was NICU admission; secondary outcomes were Apgar scores and length of infant hospital stay. To determine whether IVF or NIFT infants had a higher risk of NICU admission, we performed multivariate logistical regression adjusting for maternal age, delivery method, and gestational age. RESULTS: Of 585 singleton deliveries, there were 523 spontaneous, 47 IVF (8.0%), and 15 NIFT conceptions (2.6%). The fertility treatment women were older (32.95.4 vs. 33.64.6 vs. 39.54.3 years, p <0.001) and of a lower parity (0.71  0.94 vs. 0.470.83 vs. 0.330.53, p<0.001), but otherwise similar across race, BMI, and infant birth weight and gestational age at delivery (p >0.05). Notably, IVF infants were more likely to be delivered by cesarean section (73.9%) compared to spontaneous (40.3%) and NIFT (46.7%) infants (p<0.001). Sixty percent of NIFT infants and 46.8% of

FERTILITY & STERILITYÒ

IVF infants were admitted to the NICU compared to 28.5% of spontaneous infants (p¼0.002). Five minute Apgar scores were lowest among IVF infants (8.21.3), while spontaneous and NIFT infant scores were 8.70.8 and 8.80.4 respectively (p¼0.0059). In multivariate logistic regression analyses, NIFT infants (Odds Ratio 4.11, 95% CI 1.33-12.76) and IVF infants (OR 2.28, 95% CI 1.13-4.60) had a higher risk of NICU admission. CONCLUSIONS: Within the late preterm cohort, the IVF and NIFT infants are admitted to the NICU more often with lower Apgar scores and a longer hospital stay, despite similar gestational age and birth weights as compared to spontaneously conceived infants. Supported by: NICHD R01 HD074368. P-187 Tuesday, October 18, 2016 IMPACT OF OVULATION METHOD AND CYCLE STIMULATION ON PREGNANCY OUTCOMES: A RETROSPECTIVE COHORT STUDY OF INTRA-UTERINE INSEMINATION WITH DONOR SEMEN IN A HOSPITAL-BASED CLINIC. J. S. Lam, F. Tekpetey, K. Shepherd, A. Vilos. Obstetrics and Gynaecology, University of Western Ontario, London, ON, Canada. OBJECTIVE: To elicit the impact of ovulation method and cycle stimulation on pregnancy and live birth rates in intra-uterine insemination (IUI) candidates using donor semen. DESIGN: A retrospective cohort study of patients undergoing IUI using donor semen who attended a hospital-based fertility clinic between January 2009 and December 2014. A total of 632 cycles were evaluated from 233 patients. MATERIALS AND METHODS: Cycles were analysed based on ovulation method, natural luteinizing hormone (LH) surge (n¼280) versus ovulation trigger with hCG or OvidrelÒ (n¼320), and 32 cycles were excluded from this evaluation due to hybrid methodology. Cycles were separately analysed based on cycle stimulation, natural unmedicated (n¼222) versus medicated (n¼410) with further subdivision based on medication type (clomiphene citrate, gonadotropin only, aromatase inhibitor, gonadotropin plus GnRH antagonist). Patients were followed for clinical pregnancy rate and live birth rate per insemination overall as well as specifically for multiples. Data was analysed for statistical significance using the statistical package for the social sciences (SPSS). RESULTS: In the ovulation method analysis there was a trend towards higher clinical pregnancy rates [22.2%(71/320) versus 17.9%(50/280)] and live birth rates [19.1%(61/320) versus 16.4%(46/280)] in the trigger group relative to the natural LH surge, but the difference was not statistically significant (p¼0.22 and 0.46 respectively). In the cycle stimulation analysis there was also no statistically significant difference in clinical pregnancy rates [19.8%(81/410) versus 19.8%(44/222), p¼0.93] and live birth rates [17.3%(71/410) versus 18.0%(40/222), p¼0.91] between the medicated and unmedicated groups. On the other hand, a statistically significant increase in clinical pregnancy rates and live birth rates of multiples was observed in the trigger group relative to the LH surge [3.4%(11/320) versus 0.4%(1/280), p¼0.01 and 2.8%(9/320) versus 0.4%(1/280), p¼0.02 respectively] and in the medicated cohort relative to the unmedicated [2.9%(12/ 410) versus 0%(0/222), p¼0.01 and 2.4%(10/410) versus 0%(0/222), p¼0.02 respectively]. Subgroup analysis by medication type showed this same significant increase in pregnancy and live birth rates of multiples [3.6%(9/251) versus 0%(0/222), p<0.01 and 2.8%(7/251) versus 0%(0/ 222), p¼0.01] with gonadotropin only versus unmedicated controls. There was no statistically significant difference in the confounding factors of age, body mass index (BMI) and total motile sperm count (TMC) between patient groups in each analysis. CONCLUSIONS: This data would suggest that in the population undergoing IUI with donor semen, use of a trigger to induce ovulation and stimulation of cycles using medication do not confer any added benefit with respect to the outcomes of overall pregnancy rate and live birth rate. On the other hand, these interventions may expose the patients to a higher risk of multiple births. P-188 Tuesday, October 18, 2016 OOCYTE UTILIZATION RATE AS PREGNANCY POTENTIAL INDICATOR (A MULTICENTRIC STUDY). S. Hamamaha P. Barriere.b a ART/PGD Department, Arnaud de Villeneuve Hospital, Montpellier, France; bCHU Nantes, Nantes, France. OBJECTIVE: How many mature (MII) oocytes obtained after ICSI are required to achieve an ongoing pregnancy in different age classes of women undergoing ICSI?

e175

DESIGN: This was a multicentre retrospective audit on anonymised outcome data captured in an Excel file, then merged and analysed centrally. Overall data came from 7791 females who had 16 279 oocyte retrievals. Data was captured from ICSI only treatment cycles in which ovarian stimulation was with follitropin alfa. Data capture was from 2011 to September 2015. Additional outcome information was obtained from subsequent vitrified embryo transfer cycles carried out in the same female cohort. MATERIALS AND METHODS: Five French IVF centres having been involved in Fertility treatment for over 10 years participated. Female age was categorized into 5 groups (25-29 years; 30-34; 35-37; 38-39 and R40). Main assessment criterion was clinical pregnancy. Nonparametric LOESS method was used to estimate the relationship between pregnancy and the number of MII oocytes (on all attempts). Relationship between pregnancy at 1st attempt and female age and number of MII oocytes was estimated by logistic regression. Each cycle was considered as independent. Furthermore, over 30% of the cryopreserved embryos have not been used yet, representing residual clinical pregnancy and live birth potential not evaluated by the data. RESULTS: The mean female age was 33 5 years and 74.6% of cycles were the first attempt and only a few (2.6%) were in the rank of 5 or more. There were in total 3979 clinical pregnancies and 804 (21.5%) from subsequent vitrified-thawed cycles. The pregnancy rate increased with the number of MII oocytes (up to 20), but in women older than 38 as few high numbers of MII oocytes were retrieved, it is not possible to give an accurate estimate in these age classes. Taking into account all fresh and frozen pregnancies, the number of MII oocytes required to achieve a pregnancy stayed relatively steady below 37 years (25-29, 23.6; 30-34, 24.7; 35-37, 25.0) and then decreased in subsequent age classes: 34.8 at 38-39 years and 37.7 at 40 years or more. These numbers revealed that the pregnancy cumulative OUR remained stable between 4.2% (years 25-29) and 4.0% (years 35-37) and then decreased to 2.7% in women 40 or older. In a multiple logistic regression analysis on the first attempt, female age (p< 0.0001) and number of retrieved MII oocytes were negatively (p< 0.0001) and positively related to pregnancy, respectively. CONCLUSIONS: Our results should lead to improve couple management. Moreover, these data may provide a tool to estimate the number of metaphase II oocytes needed to achieve a clinical pregnancy and live birth in case of female fertility preservation. Supported by: An educational grant was provided by Finox AG for centralized data management and statistical analysis. P-189 Tuesday, October 18, 2016 IS YOUNGER BETTER? DONOR AGE LESS THAN 25 DOES NOT PREDICT MORE FAVORABLE OUTCOMES AFTER IN VITRO FERTILIZATION. L. A. Humphries,a,b L. E. Dodge,a,b E. B. Kennedy,a K. C. Humm,a,b M. R. Hacker,a,b D. Sakkas.c aDepartment of Obstetrics and Gynecology, Beth Israel Deaconess Medical Center, Boston, MA; bDepartment of Obstetrics, Gynecology and Reproductive Biology, Harvard Medical School, Boston, MA; cBoston IVF, Waltham, MA. OBJECTIVE: The use of donated oocytes from younger donors can circumvent age-related infertility in women of advanced maternal age with diminished ovarian reserve. Yet little is known about the outcomes of the youngest donors (<25 years). The aim of this study was to determine whether younger donor age is associated with better outcomes after in vitro fertilization (IVF) compared with older donor age. DESIGN: Retrospective cohort study. MATERIALS AND METHODS: We included all women R18 years of age who started their first oocyte donation cycle at our center from January 2002 through December 2013. Log-binomial regression was used to compare clinical pregnancy and live birth rates among the following donor age groups: <25 years, 25 to <30 years, and R30 years. Recipient, partner, and cycle characteristics were evaluated as potential confounders.

RESULTS: Median donor age was 26 years (range: 18-34), and median recipient age and partner age were both 42 years. Among fresh cycles, increasing donor age was associated with fewer oocytes retrieved (p<0.001), but not with fertilization rate (p¼0.86). For both fresh and frozen-thawed cycles, the donor age groups were similar with regard to recipient age and body mass index, as well as partner age, indication for donation, and number of prior treatment cycles. Recipients who used the youngest donors (<25 years) had lower clinical pregnancy and live birth rates compared with the older age groups for both fresh and frozen-thawed cycles, although not all of these findings reached statistical significance. Recipients who used donors age 25 to <30 years had a significantly higher clinical pregnancy rate (relative risk: 1.3; 95% confidence interval: 1.01-1.6) compared with recipients who used the youngest donors. CONCLUSIONS: Oocyte donor age <25 years was not associated with better outcomes after IVF. The optimal donor age range may be 25 to <30 years. Supported by: Harvard Catalyst, The Harvard Clinical and Translational Science Center (National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health Award UL1 TR001102) and financial contributions from Harvard University and its affiliated academic healthcare centers. P-190 Tuesday, October 18, 2016 HOW DOES ART SINGLETONS DIFFER FROM NATURALLY CONCEIVED (NC) SINGLETONS; COMPARISON OF PERINATAL DATA OF 872 ART TO 19317 (NC) SINGLETON BABIES. A. Khudhari,a R. Hemmings,b S. Phillips,c A. M. Badeghiesh,d W. Jamal.e aGynecologist/Infertility and Reproductive Endocrinology, Montreal, QC, Canada; bOb-Gyn, McGill University, Westmount, QC, Canada; c OVO Fertility, Montreal, QC, Canada; dMcgill University, Montreal, QC, Canada; eOBGYN University of Montreal, OVO Clinic, Montreal, QC, Canada. OBJECTIVE: It has been suggested that the fresh transfer of single embryo following ovarian stimulation is associated with smaller birth weight than naturally conceived singleton for a similar gestational age. Perinatal outcomes from a large Danish database (Henningsen A. et al, March 2011) showed that the birth weight of stimulated IVF singleton babies is significantly smaller than naturally conceived ones. There is limited data on the birth weight and the average gestational age at birth of modified natural in vitro-fertilization (mnIVF) conceived babies. Does (mnIVF) singleton birth weight and gestational age at delivery differ from stimulated IVF (sIVF), frozen embryo transfer (FET) and naturally conceived (NC) singleton babies? DESIGN: A retrospective cohort study chart review between the outcome of IVF cycles and spontaneous pregnancy from January 2010 to December 2014. MATERIALS AND METHODS: The gestational age and birth weight of singleton babies conceived from (246) mnIVF, (405) sIVF and (233) FET following the single embryo transfer on day 2,3 or 5 post-fertilisation at OVO fertility clinic were compared to 19,317 naturally conceived singleton babies delivered at a community hospital in Montreal. Secondary outcomes included the type of delivery and gender. RESULTS: The average gestational ages at delivery were not clinically different although statistically significant: 38.8 weeks mnIVF, 38.8 weeks sIVF, 39.2 weeks FET babies versus 39.1 of gestation for naturally conceived babies P¼0.0001. Average birth weight were comparable for mnIVF, sIVF versus NC singleton babies (3301 grams, 3263 grams versus 3353 grams) yet FET babies had significantly higher birth weights 3453 grams P¼0.0001. CONCLUSIONS: We found that the mean birth weight of mnIVF conceived singleton babies did not differ significantly from the NC or sIVF singleton babies. Yet in comparison to previous reports, FET singleton babies had higher birth weights. The average gestational age at delivery was

Clinical pregnancy and live birth per cycle relative to donor age group < 25 years

Fresh Clinical pregnancy Live birth Frozen-thawed Clinical pregnancy Live birth

e176

All cycles (n¼527)

Donor age <25 years

Donor age 25-<30 years

n[414 208 (50.2%) 180 (43.5%) n[113 36 (31.9%) 33 (29.2%)

n[153 67 (43.8%) 60 (39.2%) n[35 9 (25.7%) 9 (25.7%)

n[207 114 (55.1%) 98 (47.3%) n[57 18 (31.6%) 15 (26.3%)

ASRM Abstracts

RR [95% CI]

1.3 [1.01-1.6] 1.2 [0.95-1.5] 1.2 [0.62-2.4] 1.1 [0.52-2.2]

Donor age R30 years n[54 27 (50.0%) 22 (40.7%) n[21 9 (42.9%) 9 (42.9%)

RR [95% CI]

1.1 [0.83-1.6] 1.04 [0.71-1.5] 1.7 [0.78-3.7] 1.6 [0.73-3.4]

Vol. 106, No. 3, Supplement, September 2016