A nomogram-based model to personalize growth hormone pretreatment in assisted reproductive technique for prediction of oocyte retrieval and embryo quality

Objective: To develop and validate a predictive nomogram for oocyte retrieval and embryo quality based on personalized growth hormone (GH) pretreatment in women undergoing assisted reproductive techniques (ART) with diminished ovarian reserve (DOR). Methods: 2,000 women ≤40 years, diagnosed with DOR, were non-randomly assigned to one of four groups based upon hormonal and metabolic markers: G1 (1-month GH pretreatment followed by ART), G2 (2-month GH pretreatment), G3 (GH given only during ovarian stimulation), and G4 (standard ART without GH). 2 IU/day of GH was given to women from G1-G3, hormonal profile was repeated and dynamic stratification was done. Results: The G2 group showed significantly higher estradiol levels, more oocytes retrieved, and a higher proportion of good-quality embryos compared to G1, G3, and G4 (P < 0.001). HOMA-IR levels were elevated in the G2 group, indicating increased insulin resistance with prolonged GH exposure. Multivariate regression identified GH dose (β = 0.48, P < 0.0001) and AMH (β = 0.35, P = 0.0013) as strong positive predictors, while HOMA-IR (β = -0.22, P = 0.012) and testosterone (β = -0.29, P = 0.028) negatively impacted outcomes. The nomogram modelled through multivariate regression provides a clinically effective tool for personalization of GH pretreatment. Conclusions: The study demonstrates that GH pretreatment significantly enhances oocyte retrieval and embryo quality. A nomogram-based predictive model stratifies patients, supports the personalization of GH treatment duration, and offers a clinically useful tool toward precision reproductive medicine.