Abstract Study question Does mild ovarian stimulation in women with PCO result in higher live birth rates during subsequent FET cycles? Summary answer Mild ovarian stimulation with FSH doses <150IU did not result in higher clinical pregnancy or livebirth rates in subsequent FET. What is known already Ovarian stimulation during IVF in women with PCO is associated with an exaggerated response, ovarian hyperstimulation syndrome, poor egg to follicle ratio, low fertilisation rates and poor blastocyst conversion. Mild ovarian stimulation, often referred to as protocols with FSH doses under 150IU, is often employed to overcome these challenges. One of the perceived benefits of this approach is improved oocyte and embryo quality reflected in lower aneuploidy rates. Study design, size, duration This was a retrospective observational study looking at 99 FET between January 2011 and Jan 2021 that followed a fresh cycle in women with a pre-treatment antral follicle count of 12 + 12 or greater. Patients were identified through the antral follicle count at the pre-treatment investigation ultrasound scan. Ultrasound findings, treatment cycle details and clinical outcomes were entered prospectively into a dedicated clinic database. Data was retrieved and analysed using SPSS V25. Participants/materials, setting, methods The study was conducted in a large IVF centre. Data on women with an AFC of 12 + 12 or above, undergoing an autologous FET cycle following a fresh cycle were collected. Women were split into those receiving <150IU of FSH (Group1, n = 51) and those receiving FSH ³150 IU (Group 2, n = 48). Binary logistic regression analysis was performed to control for confounders. Live birth was the primary outcome, with biochemical and clinical pregnancy being secondary outcomes. Main results and the role of chance Women in Group 1 were younger (30.8±3.6 v 33.8±3.65, p < 0.005) but had a similar antral follicle count (38.2±11.7 v 34.2±9.1, p = 0.07). The total number of eggs collected (24.1±13.8 v 25.9±8.8, p = 0.45) and fertilisation rate (0.59±0.2 v 0.58±0.18, p = 0.77) during their fresh cycle were comparable. Women in Group 2 had a larger number of embryos suitable for cryopreservation (7.36±4.2 v 4.8±3.5, p = 0.001) In the subsequent frozen embryo replacement cycle, there was no difference in the number or quality of embryos transferred with most women having a single embryo transfer (63% v 48%, p = 0.14) and at least one top quality embryo transferred (68.6% v 81%, p = 0.15). There was a higher biochemical pregnancy rate in Group 1 (84% v 66%, p = 0.035) but with no difference in clinical pregnancy rate (53% v 44%, p = 0.37) or live birth rate (49% v 42%, p = 0.76). Live birth rates remained comparable even after controlling for age, and number and quality of embryos transferred (OR: 1.21 (95% CI 0.50–2.94). Limitations, reasons for caution This was a retrospective analysis raising the risk of allocation bias. This study was also at risk of information bias as it relied on accurate documentation of the AFC at the pre-treatment scan. Wider implications of the findings: Patients can be reassured that both stimulation protocols result in similar live birth rates in subsequent frozen embryo replacement cycles. Prospective trials using PGT-A are required to assess whether aneuploidy could account for the discrepancy in biochemical pregnancy rates in the two groups considering the subsequent comparable clinical pregnancy rates. Trial registration number Not applicable
Read full abstract