Ovarian sensitivity index-based nomogram for predicting clinical pregnancy outcomes in patients with diminished ovarian reserve undergoing in vitro fertilization or intracytoplasmic sperm injection.

Background: Predicting clinical pregnancy outcomes in patients with diminished ovarian reserve (DOR) undergoing in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI) remains challenging owing to the unique characteristics of this patient group. Therefore, this study aimed to leverage existing predictive models for pregnancy outcomes while integrating innovative strategies to develop and validate a visualization-based predictive model specifically designed for patients with DOR undergoing IVF/ICSI treatment.

Methods: This retrospective study analyzed data from 448 patients with DOR who underwent IVF/ICSI at Guangxi Zhuang Autonomous Region Reproductive Hospital from January 2019 to August 2023. We developed and internally validated a nomogram incorporating the ovarian sensitivity index (OSI), age, and controlled ovarian hyperstimulation (COH) protocol to predict clinical pregnancy outcomes. Receiver operating characteristic (ROC) analysis, univariate and least absolute shrinkage and selection operator (LASSO) regression analyses, and multivariate logistic regression were used to construct the model. The optimal cut-off value of the OSI for predicting clinical pregnancy was 1.135.

Results: Through multivariate analysis, age, OSI, and COH protocol were identified as independent predictors. The developed nomogram demonstrated good discrimination with an area under the ROC curve of 0.744, along with satisfactory calibration and clinical utility.

Conclusion: The developed nomogram can accurately predict clinical pregnancy outcomes in patients with DOR undergoing IVF/ICSI, potentially assisting clinicians in personalized counselling and improving outcomes in this challenging patient population.