Multiple evidence suggests that thyroid function is associated with polycystic ovary syndrome (PCOS), but whether thyroid function is causally related to PCOS is unclear. To investigate whether the association reflect causality, a Mendelian randomization (MR) analysis was conducted.
Single nucleotide polymorphisms (SNPs) involved in this study were acquired from The ThyroidOmics Consortium and the IEU Open Genome-wide association study (GWAS) database, respectively. In forward MR analysis, we included normal free thyroxine (FT4, n=49,269), normal thyroid-stimulating hormone (TSH, n=54,288), hypothyroidism (n=53,423) and hyperthyroidism (n=51,823) as exposure. The outcome was defined as PCOS in a sample size of 16,380,318 individuals. The exposure in the reverse MR analyses was chosen as PCOS, while the outcome consisted of the four phenotypes of thyroid function. The inverse-variance weighted (IVW) method was performed as the major analysis, supplemented by sensitivity analyses.