Predictors of Gonadotropin Efficiency Index (GEI): a regression-based model using patient and stimulation parameters in controlled ovarian stimulation cycles

Objectives. In assisted reproductive technology (ART), op­ti­mi­zing gonadotropin usage is critical for both cli­ni­cal efficacy and cost-efficiency. The Gonadotropin Ef­fi­cien­cy Index (GEI), defined as the total gonadotropin dose per grade A embryo (IU/G1), offers a novel metric to quantify stimulation efficiency. However, predictors of GEI have not been well-characterized. Our aim was to identify patient and stimulation-related parameters that significantly predict gonadotropin efficiency using a mul­ti­va­riate regression model. Materials and method. This retrospective cohort study analyzed 300 pa­tients un­der­going controlled ovarian stimulation using either re­com­bi­nant FSH (rFSH) or human menopausal go­na­do­tro­pin (hMG). Baseline variables (age, BMI, AFC, hormonal pro­file), protocol details and outcomes (total dose, number of embryos, grade A embryos) were collected. GEI was cal­cu­lated as IU/grade A embryo. Linear regression was per­formed to identify independent predictors of GEI. Results. The mean GEI was significantly lower in the rFSH group (p<0.001). AFC (β = -0.45; p<0.001) and age (β=0.36; p=0.002) were significant predictors of higher GEI. The use of hMG, lower AFC and advanced age were in­de­pen­den­tly associated with poorer gonadotropin ef­fi­ciency. A multivariate model predicted GEI with R²=0.42. Con­clu­sions. Antral follicle count and patient age are sig­ni­fi­cant predictors of gonadotropin efficiency. Predictive modeling of GEI may offer individualized stimulation planning, mi­ni­mizing drug usage and optimizing outcomes in ART cycles.