P-607 Basal serum luteinizing hormone and total testosterone levels do not impact the outcomes of in-vitro fertilization in women with polycystic ovarian syndrome (PCOS)

Abstract

Abstract Study question Do basal serum luteinizing hormone (LH) and total testosterone (TT) levels affect the outcomes of patients with PCOS undergoing in vitro fertilization? Summary answer LH and TT does not impact number of mature oocytes, fertilization rates, and high-quality blastocysts frozen, in PCOS with GnRH-antagonist protocol IVF with hyper-responses. What is known already The impact of basal serum LH and TT levels on IVF outcomes is debatable. Worldwide, there are many clinics that will postpone ovarian stimulation until a lower basal LH or TT is obtained in women with PCOS based on the assumption that outcomes will subsequently be more successful. However, hormones are released in a pulsatile fashion and single levels are unlikely to reflect patient status. There is a paucity of research addressing this concern. Study design, size, duration We completed a single university-centre retrospective cohort study comparing the outcomes of PCOS patients undergoing freeze-all GnRH antagonist IVF cycles divided into 50% groupings for basal serum LH and TT levels between the dates of January, 2013 and December, 2019. There were 76 women identified and confirmed on chart review to have a diagnosis of PCOS per the 2003 Rotterdam criteria. Participants/materials, setting, methods Patients were divided into the 50% with the lowest LH levels(1.0-7.0 IU/L) and highest LH(7.1-29.0 IU/L). The same was done for TT(0.30-1.50 vs. 1.6-5.6 nmol/L). These two groups were compared using chi-square test and t-test as indicated. Demographics included: serum estradiol, follicle stimulating hormone, thyroid stimulating hormone, prolactin, TT, free testosterone, gravida, parity, female age, male age, semen analysis parameters, antral follicle count, and duration of infertility. IRB approval was obtained(2020-5971). Data is X±SD. Main results and the role of chance When assessing based on LH, there were no differences in demographics(P > 0.05 in all cases) except for TT(1.4±1.0 vs. 1.9±0.9 nmol/L-p=0.02) and free-testosterone(0.6±0.5 vs. 0.9±0.5 nmol/L-p=0.03). The number of oocytes collected(26.8±7.4 vs. 27.3±9.6-p=0.82), mature oocytes(19.9± 6.8 vs. 20.2±7.8-p=0.88), and frozen high-quality blastocysts(7.0±4.4 vs. 8.2±5.1-p=0.25) were similar. The percent of MII that fertilized(75%±16 vs. 74%±19-p=0.78), the percent of 2PN that grew to blastocysts(47%±23 vs. 55%±27-p=0.18) and the percent of MII that grew to blastocyst(35%±18 vs. 40%±19-p=0.25) were similar. Among subjects who underwent embryo transfer, clinical pregnancy rates(33% vs. 45%-p=0.36) and live birth rates(22% vs 32%-p=0.39) were similar when comparing lowest vs. highest percentile, respectively. When comparing demographics by TT grouping, differences were previous pregnancies(0.9±1.2 vs. 0.3±0.7-p=0.02), serum TSH(1.66±0.96 vs. 2.43±1.90 nmol/L-p=0.04), and duration of infertility(2.3±2.0 vs 3.7±2.7years-p=0.04). The number of oocytes collected(27.5±7.5 vs. 28.3±9.7-p=0.73), MII oocytes(20.5±6.9 vs. 20.4±8.3-p=0.98), and frozen high-quality blastocysts(6.9±4.0 vs. 8.1±5.7-p=0.32) were similar. The percent of MII that fertilized(73%±19 vs. 74%±17-p=0.96), the percent of 2PN that grew into blastocysts(48%±23 vs. 53%±28-p=0.50) and the percent of MII that grew to blastocyst(34%±17 vs. 38%±19-p=0.36) were similar. Among subjects undergoing embryo transfer, clinical pregnancy rates(46% vs. 44%-p=0.90) and live birth rates(38% vs 26%-p=0.33) were similar when comparing lowest vs. highest percentile respectively. Limitations, reasons for caution Our study was limited to GnRH-antagonist cycles with freeze all embryos. Thus, findings may not be applicable to IVF cycles with fresh embryo transfer. It is a retrospective study and therefore is at risk for inherent biases. Wider implications of the findings Our findings provide reassurance that elevated basal serum LH and TT levels do not impact IVF outcomes in hyper-responding patients and therefore should not result in cycle cancellation or delay in those with PCOS undergoing GnRH-antagonist cycles with anticipated hyper-responses. Trial registration number N/A