Abstract
BACKGROUND: Previous Mendelian randomization (MR) studies explored causal links between polycystic ovary syndrome (PCOS) and pregnancy complications and adverse outcomes mainly in European populations, with limited generalizability. This two-sample MR study addresses this limitation. METHODS: Independent genetic instruments for PCOS were selected at genome-wide significance (P<5×10(-8)) and statistical pruned using LDTrait. East Asian (EAS)-based genome-wide association study data were analyzed for gestational diabetes mellitus (GDM), preeclampsia, intrahepatic cholestasis of pregnancy, miscarriage, preterm birth, and gestational age at birth. The inverse variance-weighted (IVW) method with Bonferroni correction served as the primary analysis, supplemented by pleiotropy and heterogeneity tests. Multivariable MR (MVMR) assessed direct effects by adjusting for confounders. RESULTS: Genetically predicted PCOS was causally linked to increased GDM risk (OR=1.203, 95% CI: 1.00-1.433), with consistent evidence across debiased IVW (OR=1.207, 95% CI: 1.008-1.444), Bayesian Weighted MR (OR=1.204, 95% CI: 1.007-1.440), Contamination Mixture (OR=1.231, 95% CI: 1.141-1.526), and distortion test (OR=1.203, 95% CI: 1.120-1.291). The strongest single nucleotide polymorphisms (SNP)-level effect was observed for rs1894116 (YAP1), followed by rs2268361 (FSHR). MVMR confirmed independence from body mass index (BMI), though the effect disappeared after adjusting for glucose and lipid traits. No causal relationships were found between PCOS and other outcomes (P>0.05). CONCLUSION: This MR study provides evidence supporting a potential causal association between genetically predicted PCOS and GDM in East Asian populations, in contrast to previous European-based findings, and highlights possible population-specific genetic differences with implications for precision risk assessment in pregnancy.