The chromosome 11 olfactory receptor gene cluster is a candidate reproductive partner-specific compatibility locus

pregnancies. Although AOA had no effect in patients with satisfactory fertilization rate, those with sub-optimal fertilization may benefit from AOA by improving also the late embryonic development. Finally, the sperm origin influenced the effect of AOA on embryonic cleavage. P-107 Tuesday, October 31, 2017 ELIMINATION OF A POLICY OF EMPIRIC ANTIBIOTIC THERAPY FOR MALE PATIENTS PRIOR TO COLLECTION FOR IVF RESULTS IN NO DIMINUTION IN FERTILIZATION RATE. S. J. Morin,a C. R. Juneau,a S. A. Neal,a R. T. Scott, Jr.,b E. J. Forman.a aIVI/RMA, Thomas Jefferson University, Basking Ridge, NJ; bREI, IVI/RMA, Thomas Jefferson University, Basking Ridge, NJ. OBJECTIVE: Multiple studies have suggested a link between leukocytospermia (LEUK) and reduced motility and decreased fertilization capacity. Given that there is often an extended period of time between initial semen analysis (SA) and collection for IVF, some practitioners fear that the absence of LEUK on initial analysis does not preclude its presence when it matters most. Others fear risk of infection at time of fertilization. As a result, many clinics give empiric antibiotics to male partners in the days leading up to oocyte retrieval. However, no data exists to support this practice. Given these concerns, this abstract compares fertilization rates in the 1 year prior to and 1 year after a single clinic discontinued empiric antibiotic therapy for males. DESIGN: Retrospective descriptive. MATERIALS AND METHODS: Prior to 2016, all male partners received empiric doxycycline or ciprofloxacin (if allergic) from the start of their partner’s stimulation cycle until day of retrieval. All empiric treatment was discontinued in January 2016. Only males with evidence of LEUK or infection on initial SA received antibiotic treatment after this policy change. All IVF cycles for one year after policy change (1/ 2016 to 12/2016) were compared against all cycles for one year prior to change (1/2015 to 12/2015). Cases using donor sperm, frozen samples, surgically retrieved sperm, and donor oocytes were excluded. The primary outcome was fertilization rate (FR). Female partners’ charts were also reviewed for hospitalization due to suspected infection after transfer. A generalized estimating equation (GEE) model was used to control for female age and patient specific correlation among oocytes from the same cohort. RESULTS: A total of 1920 and 1753 cycles met inclusion criteria in 2015 and 2016, respectively. Pooled fertilization rate was higher in 2016 than 2015 (83.5% [16598/19886] vs. 82.5% [17353/21038], p<0.01). After controlling for female age and patient specific correlations in the GEE model, antibiotic exposure was not associated with the likelihood of successful fertilization. There were no hospitalizations for suspected infection following fresh embryo transfer in either group. CONCLUSIONS: Discontinuation of empiric antibiotic therapy for male partners in the days prior to collecting a specimen for IVF had no impact on fertilization rate. Treatment should be reserved for male patients with suspected infection. Given known untoward side effects of widespread antimicrobial use, all empiric antibiotic use in IVF regimens deserve careful scrutiny. P-108 Tuesday, October 31, 2017 THE CHROMOSOME 11 OLFACTORY RECEPTOR GENE CLUSTER IS A CANDIDATE REPRODUCTIVE PARTNER-SPECIFIC COMPATIBILITY LOCUS. R. Subarana,b,c S. Munne.c aRecombine, New York, NY; bCooperSurgical, Trumbull, CT; cCooperGenomics, Livingston, NJ. OBJECTIVE: While our best understanding of genetic disease comes from single-gene traits, work in model organism shows that most phenotypes are influenced by complex gene-gene interactions. However, since most tests for detecting such interactions in humans, genome-wide, require the pairwise comparison of millions of alleles, their presence has been difficult to detect. Here, we focus on a limited panel of allele pairs to test the hypothesis that gene-gene interactions between reproductive partners influence fertility. DESIGN: Retrospective population genetics analysis.

e148

ASRM Abstracts

MATERIALS AND METHODS: In order to limit the number of pairwise allelic comparisons in our study, we developed a novel variant filtering pipeline to reduce the number alleles tested to those predicted a priori to have the strongest phenotypic effects. We then scored the two-locus genotypes for each allelic pair in a population of 6,425 healthy Swedish individuals. RESULTS: We found a > 95% depletion (p-val ¼ 5.7 x 10-85), compared to expected, of the double homozygous non-reference alleles of rs1459101 and rs11228710 at OR4C16 and OR5AR1. While these alleles are syntenic, they are separated by 1.09 Mb and are not part of a contiguous linkage disequilibrium block in our sample. Furthermore, each of these alleles has a frequency > 10% in multiple major populations, making it unlikely that our observation is the result of admixture based on geographic location. CONCLUSIONS: Our result suggests that the reason for observed twolocus genotype depletion is that individuals homozygous for the minor allele at one locus reproduce less frequently with individuals of homozygous for the minor allele at the other. One explanation is that OR4C16 and OR5AR1 play some direct role in reproduction. Many olfactory receptors are highly expressed in mammalian spermatozoa, where they have been proposed as having a putative role in chemotaxis. However, that alone would not explain allele-specific incompatibility with specific female olfactory receptor genotypes. Perhaps the most straightforward explanation is that our findings are the result of some previously described molecular mechanism for the scent cue-based mate selection observed in humans and in model mammalian organisms. Whether this will be borne out bears further investigation.

P-109 Tuesday, October 31, 2017 INCREASING BMI IS ASSOCIATED WITH DECREASED FERTILIZATION RATE AND INCREASED FERTILIZATION FAILURE IN ART a a CYCLES. P. Bortoletto, A. M. Lee, J. C. Vanegas,b E. Hariton,a M. Palmor,a L. A. Humphries,a J. E. Chavarro,b A. K. Styer.a aMassachusetts General Hospital Fertility Center, Boston, MA; bHarvard T.H. Chan School of Public Health, Boston, MA. OBJECTIVE: To evaluate the association of body mass index (BMI) and fertilization rates following controlled ovarian hyperstimulation for IVF. DESIGN: Retrospective cohort. MATERIALS AND METHODS: 839 consecutive IVF cycles (first attempt) utilizing autologous oocyte in patients without a diagnosis of male factor infertility from 2006-2014 were examined. A logistic regression model to estimate the association of BMI and normal fertilization rate (the percent at two pronuclear stage per M2 oocyte 24 hours after IVF) and poor fertilization rate (defined as <30% normal fertilization) per cycle initiated was fitted while adjusting for age, infertility diagnosis, stimulation protocol, and insemination type (IVF or elective ICSI [ICSI]). Patients were classified into four categories of BMI (kg/ m2): underweight (BMI < 18.5), normal weight (BMI 18.5-24.9), overweight (BMI 25-29.9) and obese (BMI R 30). Normal weight patients were further divided into low normal (BMI 18.5-22.4) and high normal (BMI 22.5-24.9) BMI. RESULTS: There were 22 (2.6%) underweight, 528 (62.9%) normal weight, 202 (24.1%) overweight, and 87 (10.4%) obese women. Age, ethnicity, duration of infertility, treatment protocol, and type of insemination was similar among groups. Obese women required the most gonadotropins (P¼0.001) and had the lowest peak estradiol levels (P¼0.001). Following multivariable-adjusted analyses, a linear decrease in the rate of normal fertilization was observed with increasing BMI. Fertilization rates of 74.8% (underweight), 74% (low normal BMI), 69.4% (high normal BMI), 70.7% (overweight), and 68.8% (obese) (P, linear trend <0.001) were observed, This association was less pronounced in ICSI cycles compared to IVF cycles (P, linear trend 0.63). Specifically, a 1 kg/m2 increase in BMI was associated with a 2.3 % (95% CI: 1.6 - 3.8) reduction in fertilization rate in all (IVF and ICSI) cycles, a 3.2 % decrease (95% CI: 1.1 - 4.3) in IVF cycles, and a 1.2% (95% CI: -2.2 - 4.1) decrease in ICSI cycles respectively. With every 1 kg/m2 increase in BMI, a 5% increased risk of fertilization failure was observed (OR: 1.05; 95% CI 1.0 to 1.1; P ¼ 0.04). CONCLUSIONS: As BMI increases, the rate of normal fertilization decreases and the risk of failed fertilization increases. Since this relationship was attenuated with ICSI, there may be possible utility for ICSI to optimize

Vol. 108, No. 3, Supplement, September 2017