Principal Findings
This 13-year retrospective cohort study revealed an increased incidence of Graves’ diseases in PCOS patients with an aHR of 1.34 (95%C.I. 1.00-1.81) after adjusting for age and confounders listed in table 1. In stratification analysis, the independent association of PCOS with Graves’ disease was especially significant in group aged 31 to 40 (aHR=2.31, 95%C.I. 1.48-3.61). The cluster of PCOS and Graves’ disease at this age may due to the nature of PCOS, a common endocrinopathy of reproductive age-group women. Several comorbidities were found influential in our Cox regression model, such as CAD (aHR=2.66, 95%C.I. 1.08-6.56) and DM (aHR=2.17, 95%C.I. 1.06-4.44). In our stratification study investigating into the association of PCOS and Graves’ diseases in different subgroups, we found that in patients with hypertension, the independent association of PCOS with Graves’ disease became insignificant, which may be explainable due to probable effect of other coexisting comorbidities and relatively small numbers of events. However, in patients with hyperlipidemia, PCOS increases the risk of Graves’ diseases by 10 folds, higher than PCOS acting alone. In general, after adjusting for confounders in our study, the overall independent association of PCOS with Graves’ disease remains strong, which may supports the hypothesis of PCOS as a possible risk factor to the development of Graves’ diseases.