Endocrine connections · 2026
Development and external validation of a parsimonious endocrine-metabolic model for 12-month pregnancy prediction in women with polycystic ovary syndrome (PCOS).
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Ha Sifu, Liu Lina, Li Jian, Yu Haixin, Liu Hong, Ma Lin, Liu Xin
Аннотация
BACKGROUND: Women with polycystic ovary syndrome (PCOS) exhibit diverse reproductive and metabolic phenotypes, complicating short-term pregnancy counseling. We developed and externally validated a pragmatic model to predict clinical pregnancy within 12 months.
METHODS: This retrospective cohort study utilized data from Center A (development) and Center B (validation). Predictors included routinely available anthropometric, reproductive endocrine, metabolic, and treatment markers. We developed a multivariable logistic regression model, evaluating discrimination (AUC and precision-recall), calibration (decile-based plots), and clinical utility via decision curve analysis (DCA). Clinical pregnancy, confirmed by ultrasonography, served as the reference standard. A web-based calculator was created for bedside risk estimation.
RESULTS: In the development cohort (n = 315; 167 pregnancies), the final model retained age, anti-Müllerian hormone (AMH), LH/FSH ratio, assisted reproductive treatment, HOMA-IR, and homocysteine. The model demonstrated good discrimination (AUC = 0.813) and remained robust after internal validation (AUC = 0.800; AUPRC = 0.823). Calibration showed strong agreement between predicted and observed probabilities, despite minor deviations at higher risk levels. DCA indicated higher net benefit compared to 'treat-all' or 'treat-none' strategies across relevant thresholds. External validation confirmed acceptable discrimination (AUC = 0.759), with modest calibration attenuation (slope = 0.681; intercept = -0.019).
CONCLUSION: This endocrine-metabolic model provides individualized 12-month pregnancy predictions for women with PCOS, demonstrating validated discrimination and decision-analytic utility. While the web calculator facilitates clinical implementation, local recalibration is recommended to ensure accuracy across different populations.
METHODS: This retrospective cohort study utilized data from Center A (development) and Center B (validation). Predictors included routinely available anthropometric, reproductive endocrine, metabolic, and treatment markers. We developed a multivariable logistic regression model, evaluating discrimination (AUC and precision-recall), calibration (decile-based plots), and clinical utility via decision curve analysis (DCA). Clinical pregnancy, confirmed by ultrasonography, served as the reference standard. A web-based calculator was created for bedside risk estimation.
RESULTS: In the development cohort (n = 315; 167 pregnancies), the final model retained age, anti-Müllerian hormone (AMH), LH/FSH ratio, assisted reproductive treatment, HOMA-IR, and homocysteine. The model demonstrated good discrimination (AUC = 0.813) and remained robust after internal validation (AUC = 0.800; AUPRC = 0.823). Calibration showed strong agreement between predicted and observed probabilities, despite minor deviations at higher risk levels. DCA indicated higher net benefit compared to 'treat-all' or 'treat-none' strategies across relevant thresholds. External validation confirmed acceptable discrimination (AUC = 0.759), with modest calibration attenuation (slope = 0.681; intercept = -0.019).
CONCLUSION: This endocrine-metabolic model provides individualized 12-month pregnancy predictions for women with PCOS, demonstrating validated discrimination and decision-analytic utility. While the web calculator facilitates clinical implementation, local recalibration is recommended to ensure accuracy across different populations.