Initial ovarian sensitivity index predicts embryo quality and pregnancy potential in the first days of controlled ovarian stimulation

Abstract

Abstract Background To determine if a modified ovarian sensitivity index (MOSI), based on initial follicular measurements and the initial follicle-stimulating hormone (FSH) dose, can predict the production of high-quality embryos for successful implantation during in vitro fertilization (IVF). Methods This study consisted of two phases: 1) a retrospective study and 2) a prospective observational study. For the first phase, 363 patients charts were reviewed, of which 283 had embryos transferred. All women underwent a standardized antagonist-based IVF protocol. At the first follow-up (Day 3/4), the number and size of the follicles were determined. MOSI was calculated as ln (number follicles (≥6 mm) × 1000 / FSH initial dose). Afterward, the number and quality of the ova, embryo development, and the number and quality of the blastocysts were determined. Embryo implantation was confirmed by β-hCG. For the second phase, 337 IVF cycles were followed to determine MOSI’s accuracy. Results MOSI could predict the production of ≥4 high-quality embryos by Day 2 (AUC = 0.69, 95%CI:0.63–0.75), ≥2 blastocysts (AUC = 0.74, 95%CI:0.68–0.79), and ≥ 35% rate of blastocyst formation (AUC = 0.65, 95%CI:0.58–0.72). Using linear regression, MOSI was highly associated with the number of ova captured (β = 5.15), MII oocytes (β = 4.31), embryos produced (β = 2.90), high-quality embryos (β = 0.98), and the blastocyst formation rate (β = 0.06, p < 0.01). Using logistic regression, MOSI was highly associated with achieving ≥4 high-quality embryos (odds ratio = 2.80, 95%CI:1.90–4.13), ≥2 blastocysts (odds ratio = 3.40, 95%CI:2.33–4.95), and ≥ 35% blastocysts formation rate (odds ratio = 1.96, 95%CI:1.31–2.92). This effect was independent of age, BMI, and antral follicle count. For implantation, MOSI was significantly associated with successful implantation (odds ratio = 1.79, 95%CI:1.25–2.57). For the prospective study, MOSI was highly accurate at predicting ≥6 high-quality embryos on Day 2 (accuracy = 68.5%), ≥6 blastocysts (accuracy = 68.0%), and a blastocyst formation rate of ≥35% (accuracy = 61.4%). Conclusion MOSI was highly correlated with key IVF parameters that are associated with achieved pregnancy. Using this index with antagonist cycles, clinicians may opt to stop an IVF cycle, under the assumption that the cycle will fail to produce good blastocysts, preventing wasting the patient’s resources and time.