Development and validation of a nomogram for predicting the probability of live birth in infertile women

Abstract

Objective: To develop a nomogram to predict the probability of live birth on the basis of the association of patient characteristics in subfertile individuals or couples. Methods: A retrospective study was conducted from January 2014 to December 2015. A nomogram was built from a training cohort and tested on an independent validation cohort. A total of 2,257 patients who had undergone their first nondonor cycle of in vitro fertilization (IVF) (including intracytoplasmic sperm injection) were randomly split 2:1 into training (n = 1,527) and validation (n = 730) cohorts. Results: There were no statistically significant differences in the patients' baseline and cycle characteristics between the training and validation cohorts. On multiple logistic regression analysis, female age, antral follicle count, tubal factor, anovulation, ethnicity, unexplained fertility, and male factor were significantly associated with live birth. The nomogram had a C-index of 0.700 (95% confidence interval [CI]: 0.698–0.701) in the training cohort and 0.684 (95% CI: 0.681–0.687) in the validation cohort. Conclusions: Our nomogram can predict the probability of live birth for infertile women and can be used to guide clinicians and couples to decide on an IVF treatment option.

Keywords

• In vitro Fertilization/Intracytoplasmic Sperm Injection; Live Birth Rate; Nomogram; Predictive Model; Treatment Outcome Prediction