P-449 Association of corpus luteum activity in early pregnancy and live birth after assisted reproductive technology (ART) and intrauterine insemination (IUI)

Abstract

Abstract Study question What is the incidence of low corpus luteum activity in early pregnancy, and how are endogenous progesterone levels and live birth achievement associated? Summary answer Frequently occurring low corpus luteum activity (progesterone-level) in early pregnancy after ART and IUI strongly predicts pregnancy demise, though not independent of trophoblast activity (hCG-level). What is known already Ovarian stimulation and human chorion gonadotropine (hCG) administration may induce a luteal phase defect via negative feedback on the hypothalamic-pituitary-ovarian (HPO) axis. Luteal phase support (LPS) is routinely administered with bio-identical progesterone or oral dydrogesterone (DYD). DYD has high oral availability, minimal interaction with the pituitary gland, and no cross-reactivity in enzyme-linked immunosorbent assay (ELISA) with progesterone. Therefore, utilizing DYD as LPS in IVF/ICSI or ovulatory IUI cycles provides a unique perspective to assess endocrine luteal activity. While various studies explored progesterone and hCG post-IVF or IUI, none focused solely on the corpus luteum’s sole endocrine function and its outcomes. Study design, size, duration This single-center, retrospective study (with institutional review board approval) investigated hCG and progesterone serum levels after IVF/ICSI cycles or IUI cycles supported with DYD as the sole LPS. The study aims to assess their predictive value, individually and in combination, for sustaining pregnancy. Data from 189 IVF/ICSI and 198 IUI early pregnancies, respectively, with positive hCG test between January 2017 and June 2022, was extracted from the patient management system at the University Fertility Center. Participants/materials, setting, methods Included were only first IVF/ICSI and IUI treatment-cycles with serum hCG value ≥ limit of detection (0.6 IU/l) between days 9 and 16 after fresh embryo transfer or IUI, respectively. Additionally to hCG, endogenous progesterone was determined and the relationship between these and pregnancy maintenance was investigated using logistic regression analysis. Inclusion criteria encompassed hCG-triggered cycles using DYD as the sole LPS, administered at 30 and 20 mg daily for IVF/ICSI and IUI therapy, respectively. Main results and the role of chance The 10th, 25th, 50th and 75th quantiles of serum progesterone are at 0.28, 3.75, 33.0 and 98.7 µg/l in IVF/ICSI cycles and 0.48, 2.70, 18.40 and 26.6 µg/l in IUI cycles. Women in the first quartile of progesterone levels had nearly exclusively pregnancy demise: live birth rate 3/47 (7%, 95% confidence interval (CI) 1-19%) and 0/47 (0%, 95% CI 0-7.5%) in IVF/ICSI and IUI cycles, respectively. In both IVF/ICSI and IUI, progesterone and hCG were highly correlated (correlation coefficient 0.83 in both cases). Accordingly, the association of serum hCG levels with live birth achievement was similar in quartile analyses. Using receiver operating characteristic (ROC) curves and Youden’s indices, the optimal cut-off values for predicting live birth were determined as follows: 16.49 µg/l (89 % sensitivity (SN) and 59 % specificity (SP)) for IVF/ICSI and 15.89 µg/l (90 % SN and 70 % SP) for IUI therapy. Logarithmic regression analysis also revealed a significant correlation between progesterone levels and live birth in IVF/ICSI and IUI therapy, with a p-value of < 0.001. Limitations, reasons for caution Low progesterone in early pregnancy may be a consequence or a cause for pregnancy demise. DYD, like any other progesterone for LPS, might adversely impact corpus luteum formation and maintenance through HPO axis interaction, especially in mono-ovulatory cycles or in older women with inherently higher risk of luteal phase defect. Wider implications of the findings Ovarian stimulation and hCG-administration may induce a luteal phase defect with insufficient luteal activity in early pregnancy, even up to luteolysis, with potential consequences for pregnancy maintenance and maternal/fetal health. Future treatment protocols may require exploring an optimal luteotropic state at pregnancy’s outset, necessitating the establishment of relevant markers. Trial registration number not applicable