To determine the optimal luteinising hormone (LH) level on the trigger day and its impact on pregnancy outcomes in gonadotropin-releasing hormone (GnRH) antagonist protocols using a data-driven approach. Retrospective cohort study. Third Affiliated Hospital of Guangzhou Medical University. Overall, 6107 invitro fertilisation/intra-cytoplasmic sperm injection fresh embryo transfer cycles with GnRH-antagonist protocols were performed between January 1, 2018 and February 1, 2023. Restricted cubic spline analysis and segmented regression identified the optimal LH threshold. Patients were categorised into low (≤ 1.6 IU/L) and high (> 1.6 IU/L) LH groups. Propensity score matching (PSM) and multivariable logistic regression were applied to adjust for confounding factors. Live birth rate per embryo transfer cycle. The high LH group showed significantly higher live birth rates (42.9% vs. 36.9%, adjusted odds ratio [aOR]: 1.468; 95% CI: 1.220-1.766, p < 0.001), ongoing pregnancy rates (51.4% vs. 43.6%, aOR: 1.498; 95% CI: 1.338-1.678, p < 0.001), clinical pregnancy rates (52.4% vs. 45.6%, aOR: 1.439; 95% CI: 1.285-1.611, p < 0.001) and biochemical pregnancy rates compared with the low LH group, despite retrieving fewer oocytes (median 10 vs. 12, p < 0.001). These results remained consistent after PSM and multivariable logistic regression analysis. Higher LH levels on the trigger day are associated with improved pregnancy outcomes in GnRH-antagonist protocols. Maintaining an optimal LH range is crucial for balancing oocyte yield and assisted reproductive technology success, highlighting the importance of individualised ovarian stimulation protocols.
Read full abstract