High singleton live birth rate confirmed after ovulation induction in women with anovulatory polycystic ovary syndrome: validation of a prediction model for clinical practice

Abstract

To evaluate the cumulative singleton live birth rate after classic ovulation induction in women with anovulatory polycystic ovary syndrome and to validate a previously developed prediction model. Prospective follow-up study. Tertiary infertility unit. Validation cohort of 108 treatment-naïve anovulatory PCOS patients. Conventional ovulation induction, applying clomiphene citrate as first-line treatment, followed by exogenous gonadotropins as second-line intervention. Singleton live birth prediction. Model calibration and discrimination were assessed for the initial model (variables included age, duration of infertility, and insulin/glucose ratio) and a second model in which the insulin/glucose ratio was replaced by body mass index. The cumulative singleton live birth rate after 12 and 24 months was 60% and 78%, respectively. Overall, the observed rates were higher than predicted: hazard ratio 1.21 (95% confidence interval [CI] 0.89, 1.64), first model and 1.25 (95% CI 1.20, 1.30), second model. However, the predictive capacity of the model variables was reliable, with calibration slopes of 0.79 (95% CI -0.04, 1.63) and 1.06 (95% CI 0.95, 1.18), respectively. The present study confirms the previously reported good treatment prognosis for women with PCOS undergoing classic ovulation induction. Women with a poor prognosis, for whom alternative treatment options may be considered, can best be identified by a prediction model including age, duration of infertility, and body mass index. NCT00821379.