P-633 Wednesday, October 16, 2013

THE RELATIONSHIP BETWEEN SEX HORMONE BINDINGGLOBULIN AND CONTROLLED OVARIAN HORMONAL STIMU-LATION IN ASSISTED REPRODUCTIVE TECHNOLOGYPATIENTS. L. L. Penrose,a B. Mosman,b S. D. Prien.a aDepartment ofObstetrics and Gynecology, Texas Tech University Health Sciences Center,Lubbock, TX; bDepartment of Animal and Food Sciences, Texas Tech Uni-versity, Lubbock, TX.

OBJECTIVE: Sex Hormone Binding Globulin (SHBG) has a predictablepattern during natural cycles. However, a previous study suggested thatSHBG, determined using a radioimmunoassay (RIA), increased with stimu-lation procedures during ART. Recently a more sensitive, automated assaywas developed for SHBG. The objective of the current study was to reex-amine the relationship between SHBG and assisted reproductive technology(ART) cycles, and attempt to determine how SHBGmight be associated withcirculating hormone levels and cycle outcomes.

DESIGN: Laboratory based study of SHBG/hormone relationships andcycle outcomes.

MATERIALS AND METHODS: Serum was obtained from all patientswho had undergone ARTover a one-year period.While stimulation protocolsvaried, all patients had undergone controlled ovarian hormonal stimulation(COH) and received progesterone support. Serum hormone levels for estra-diol, progesterone, testosterone, as well as SHBG were obtained from threedates: 1) initiated COH, 2) within two days of retrieval and 3) at transfer.Further, data were collected on cycle outcome. Data were then compared us-ing regression or ANOVA.

RESULTS: Data from the new SHBG assay were higher than those previ-ously reported in the literature for COH patients. Further, all patients demon-strated a drop in SHBG at the time of retrieval (132.4 +/- 67.5 nmol/L)compared to the date of COH start (205.4 +/- 90.6), which rebounded bythe date of transfer (207.3 +/- 98.6; P < 0 .001). While the initial SHBG’swere almost 20% lower in patients who conceived vs. those who did not(185.4 vs 224.3) they were statistically similar (P ¼ 0.213). Further the ratioof SHBG to the three steroid hormones remained similar throughout the ARTcycle (P ¼ 0.224).

CONCLUSION: The data suggest the new SHBG assay might be moresensitive that the RIA method previously reported. While the data demon-strate a new pattern of SHBG in COH patients not previously described inthe literature, there appeared to be no relationship between SHBG and preg-nancy outcome.

P-634 Wednesday, October 16, 2013

ESTRADIOL ON THE SECOND DAY IS NOT RELATED TO PREG-NANCY RATE. M. C. R. M. Albuquerque,a A. C. Azevedo,a

F. C. Miyasato,b T. V. Oliveira,a D. Souza,c L. E. T. Albuquerque.d aIVFLab, Fertivitro - Human Reproduction Center, Sao Paulo, SP, Brazil; bMed-ical, Fertivitro - Human Reproduction Center, Sao Paulo, SP, Brazil; cNurs-ing, Fertivitro - Human Reproduction Center, Sao Paulo, SP, Brazil; dClinicalDirector, Fertivitro - Human Reproduction Center, Sao Paulo, SP, Brazil.

OBJECTIVE: The real importance of estradiol levels on the second day ofthe menstrual cycle in IVF results remains controversial. Up to now, there arefew studies in the literature and there is no consensus among the authors ifhigh level of estradiol on the beginning of the cycle reduces the chances ofpregnancy. The importance to know if this relationship exist is in fact, is toanalyze the possibility to start or cancel an IVF cycle when estradiol is above75 ng / mL. This study aims to assess if the levels of serum estradiol on thesecond day of the menstrual cycle is related with pregnancy rates in IVF cy-cles.

DESIGN: We retrospectively analyzed 102 patients undergoing IVF tech-nique at the Human Reproduction Center FERTIVITRO, from June 2011 toJune 2012, where blood samples were collected on the second day of themenstrual cycle in order to evaluate serum levels of estradiol.

MATERIALSANDMETHODS: The groups were stratified by age. GroupI:% 37 years. Group II:R 38 years. The blood samples of all patients werecollected and the estradiol levels were analyzed and compared between preg-nant and non pregnant patients of each group.

RESULTS: In Group I the estradiol concentration average was 51.67 ng /ml (� 16.43), and in the pregnants patients was 51.56 ng / ml (� 16.5). InGroup II, the estradiol concentration average was 49.21 ng / ml (� 22.14)and in the pregnant patients was 48.29 ng / ml (� 17.68). The abortionrate was 25.6% when the estradiol concentration average was 48.35 ng /ml (� 6.05) and the average age was 36 years (� 5.4).

FERTILITY & STERILITY�

CONCLUSION: Our results showed no statistical difference on the serumlevels of estradiol in pregnant and non pregnant patients. Due to this, we canbelieve that the plasma estradiol level on the second day of the menstrual cy-cle does not predict success or failure in attempts of IVF and also we notobserved relationship with the abortion rates.

P-635 Wednesday, October 16, 2013

IN VIVO ESTRADIOL EFFECTS ON VEGF RECEPTOR 1 EXPRES-SION IN HUMAN ENDOMETRIUM: A RANDOMIZED,CONTROLLED STUDY. W. B. Davenport, S. Hale, P. Casson,I. Bernstein. Department of OBGYN, University of Vermont, Burlington,VT.

OBJECTIVE: Vascular endothelial growth factor receptor 2 (VEGFR2)induces angiogenesis and permeability in human adult endometrium by apathway that is primarily initiated by VEGF. However, VEGF receptor 1(VEGFR1) is thought to compete for substrate, attenuating VEGFR2 activity.Our recent research in humans has suggested that VEGFR2 is dependentupon estradiol(E2) in human endometrium, but no studies have observedthe effects of in vivo E2 on the expression of VEGFR1 protein in humanendometrium under controlled hormonal exposure, our study’s objective.DESIGN: Randomized Controlled Study.MATERIALS AND METHODS: Endometrial biopsies were collected

from participants during the proliferative phase of a spontaneous cycle onday 11 (CD11). Each participant then underwent ovarian suppression therapywith leuprolide depot (3.75mg IM) for a minimum of 10 days. Following sup-pression, each participant was randomized to receive either 1)transcutaneousE replacement to mimic spontaneous cycle serum E, or 2)placebo. SuccessfulE suppression/replacement was confirmed. Endometrial biopsies and serumE levels were again performed 2 and 11 days after randomization. Immuno-blotting, followed by densitometry, for VEGFR1was performed on the endo-metrial biopsy lysates using GAPDH as control. ANOVA was used todetermine significance. Data is expressed as mean�SE, p< .05 was acceptedfor significance.RESULTS: 7 participants were studied, 4 randomized to E replacement

and 3 to placebo. Women were age 27+/-4 y, healthy and cycling regularly,and of normal BMI (24+/-3kg/m2). The two groups did not differ in age orBMI. Eleven days of E replacement increased VEGFR1 signal intensity to3-fold that of the Placebo group (p¼ .04) and levels of VEGFR1 were com-parable between spontaneous CD 11 and E day 11 (p¼ .80). Within the Egroup, VEGFR1 signal was 4-fold higher on E day 11 vs E day 2 (p¼ .01).CONCLUSION: Our study suggests that VEGFR1 expression is at least

partially affected by E2 levels, both physiologically and after ovarian sup-pression.Supported by: Ferring Grant.

P-636 Wednesday, October 16, 2013

THE RELATIONSHIP BETWEEN THE NUMBER OF CGG RE-PEATS AND SERUM LEVEL OF ANTI-M€ULLERIAN HORMONEIN WOMEN WITHOUT FMR1 PREMUTATION. S. K. Kim,a

B. C. Jee,a,b S. H. Kim,b Y. M. Choi.b aObstetrics & Gynecology, Seoul Na-tional University Bundang Hospital, Seongnam-si, Gyeonggi-do, Republicof Korea; bSeoul National University Colledge of Medicine, Seoul, Republicof Korea.

OBJECTIVE: To investigate the relationship between the number of CGGrepeats in fragile Xmental retardation 1(FMR1) gene and serum level of anti-M€ullerian hormone (AMH) in Korean infertility patients without premuta-tion.DESIGN: This is a retrospective study.MATERIALS AND METHODS: We reviewed 228 infertile women who

received fertility treatment in a single private in vitro fertilization (IVF) clinicfrom May 2010 to August 2012. Serum FSH and AMH was measured onmenstrual day 3 and the number of CGG repeats was evaluated as indicated.RESULTS: The mean age of the study population was 33.3�3.8 years. No

significant correlation was observed between CGG repeat count in both al-leles and the serum FSH, AMH or multiples of median (MoM) of AMH inwhole study subjects. However, in women with age R 35 years, there wasan increasing tendency in the MoM of AMH with increasing number ofCGG repeats in allele 2 (r2¼0.075, p¼0.008). This correlation was notobserved in patients aged less than 35 years.CONCLUSION: We observed a positive correlation between MoM of

AMH and number of CGG repeats in allele 2 in women aged over 35 years.

S331