Office hysteroscopy in pre- and post-menopausal women: a predictive model

Objectives To assess the variables associated with success of office hysteroscopy (OH) in pre-menopausal and post-menopausal women and to develop a clinical model for predicting the outcome of OH. Methods This is a retrospective cohort study of consecutive patients (n = 3181) referred for an OH to a tertiary care university hospital between January 2018 and March 2020. Multivariate logistic regression analysis was used to investigate the variables for predicting the success of OH in all patients and in pre-menopausal and in post-menopausal patients separately. The logistic regression analysis of each variable was applied to develop a predictive model. Results The overall success rate of the procedure was 92.2%; 95.4% in pre-menopausal women and 87.6% in post-menopausal women. In the general population, independent predictors of procedure success were previous vaginally delivery and hysteroscopy, while previous cervical or uterine surgery were associated with incomplete OH. In the pre-menopausal group, the independent predictors of failure were treatment with GnRH, estroprogestins and infertility. In 89% of cases, our developed model was able to predict whether an OH would be successful in a particular patient. ROC analysis showed an area under the curve of 0.8746 (95% CI: 0.85354-0.89557). Conclusions The present study demonstrates the development of a simple and reliable clinical model for the identification of both pre-menopausal and menopausal patients with a high chance of OH success.