Association of pre-pregnancy abnormal glucose metabolism and insulin resistance with assisted reproductive technology outcomes and pregnancy complications in women with PCOS: a retrospective propensity score matching study.

内容摘要

Polycystic ovary syndrome (PCOS) is frequently accompanied by insulin resistance (IR) and abnormal glucose metabolism, impairing fertility and pregnancy outcomes. This retrospective study aimed to investigate their associations with assisted reproductive technology (ART) outcomes and pregnancy complications in women with PCOS. This retrospective study enrolled 1815 women with PCOS who underwent ART at Renmin Hospital of Wuhan University from January 2019 to January 2025. Patients were stratified into two groups according to pre-pregnancy glucose and insulin status: a study group with abnormal glucose metabolism or IR, and a control group with neither condition. Propensity score matching (PSM) was performed to balance baseline characteristics. Multivariate logistic regression was used to screen factors associated with live birth, preterm birth, gestational diabetes mellitus (GDM), and pregnancy-induced hypertension (PIH). A multivariate exploratory prediction model and receiver operating characteristic (ROC) curve were constructed to preliminarily evaluate the predictive efficacy and discriminative performance for GDM and PIH. After PSM, the study group had a significantly lower live birth rate and a higher miscarriage rate than the control group (both P&#x2009;<&#x2009;0.001). Pre-pregnancy abnormal glucose metabolism/IR emerged as an independent associated factor of reduced live birth (OR&#x2009;=&#x2009;0.551). For preterm birth, female age (OR&#x2009;=&#x2009;0.843), GDM (OR&#x2009;=&#x2009;2.789), and the number of embryos transferred (OR&#x2009;=&#x2009;2.174) were identified as independent associated factors. Female age (OR&#x2009;=&#x2009;1.305) and fasting plasma glucose (FPG) (OR&#x2009;=&#x2009;1.865) were significant associated indicators of GDM, while body mass index (BMI) (OR&#x2009;=&#x2009;1.506), embryo transfer type (OR&#x2009;=&#x2009;6.406), and FPG (OR&#x2009;=&#x2009;2.224) were associated with PIH. Two preliminary exploratory multivariate prediction models for GDM (AUC&#x2009;=&#x2009;0.746 pre-PSM, 0.712 post-PSM) and PIH (AUC&#x2009;=&#x2009;0.741 pre-PSM, 0.711 post-PSM) showed moderate cohort-specific predictive efficacy for the respective complications. Pre-pregnancy abnormal glucose metabolism or IR is independently associated with adverse ART and pregnancy outcomes in women with PCOS, including increased miscarriage risk, reduced live birth rate, and elevated odds of GDM. We identified distinct independent associated factors for live birth, preterm birth, GDM, and PIH. Furthermore, we developed preliminary exploratory multivariate prediction models for GDM and PIH. These models demonstrated moderate discriminative ability in our cohort but require external validation before any clinical application can be considered.