Combined metabolic-reproductive association and predictive value of AMH and TyG index in PCOS: a single-center retrospective study.
Polycystic ovary syndrome (PCOS) is a heterogeneous endocrine disorder associated with both reproductive and metabolic dysregulation. While anti-Müllerian hormone (AMH) and the triglyceride-glucose (TyG) index have been independently linked to PCOS, their combined and interactive effects remain unclear. This study aimed primarily to investigate the independent and synergistic associations of AMH and the TyG index with PCOS, and secondarily to evaluate their predictive performance for PCOS risk.
A single-center retrospective cross-sectional study included 628 participants (2020-2024). Clinical, hormonal, and metabolic parameters were assessed. Associations and predictive ability were examined using logistic regression restricted cubic splines (RCS), subgroup analysis, and receiver operating characteristic (ROC) curves.
Adjusted analyses confirmed AMH and the TyG index as independent risk factors for PCOS (ORs = 1.252, 1.981; both P < 0.05). A significant interaction was identified (OR = 1.029, P < 0.001). In subgroup analyses, a significant positive association was observed between the TyG index and PCOS risk in the low AMH subgroup (OR = 4.561, P = 0.002). Compared with the double-low group (low AMH and low TyG), the risk for PCOS showed a sequential increase in the low AMH/high TyG, high AMH/low TyG, and double-high groups (ORs = 3.582, 5.894, and 12.082; all P < 0.05), demonstrating a clear dose-response relationship. The predictive model combining AMH and the TyG index demonstrated the highest accuracy (AUC = 0.693), which was slightly higher than that of either indicator alone.
Beyond their independent roles, AMH and the TyG index interact synergistically in the pathogenesis of PCOS.
A single-center retrospective cross-sectional study included 628 participants (2020-2024). Clinical, hormonal, and metabolic parameters were assessed. Associations and predictive ability were examined using logistic regression restricted cubic splines (RCS), subgroup analysis, and receiver operating characteristic (ROC) curves.
Adjusted analyses confirmed AMH and the TyG index as independent risk factors for PCOS (ORs = 1.252, 1.981; both P < 0.05). A significant interaction was identified (OR = 1.029, P < 0.001). In subgroup analyses, a significant positive association was observed between the TyG index and PCOS risk in the low AMH subgroup (OR = 4.561, P = 0.002). Compared with the double-low group (low AMH and low TyG), the risk for PCOS showed a sequential increase in the low AMH/high TyG, high AMH/low TyG, and double-high groups (ORs = 3.582, 5.894, and 12.082; all P < 0.05), demonstrating a clear dose-response relationship. The predictive model combining AMH and the TyG index demonstrated the highest accuracy (AUC = 0.693), which was slightly higher than that of either indicator alone.
Beyond their independent roles, AMH and the TyG index interact synergistically in the pathogenesis of PCOS.