Thyroid nodules (TNs) are highly prevalent worldwide and have a pattern of female predominance. Bisphenol A (BPA) is an endocrine disruptor that can lead to adverse effects in human health. However, epidemiologic studies revealing the association between BPA exposure and TNs are limited and the results are inconsistent. We aimed to examine the association between urinary BPA and TNs in women who are more susceptible to TNs.We conducted a case-control study with 1416 women aged 18 years or older (705 cases, 711 controls). All participants underwent thyroid ultrasonography. Urinary total BPA (free and conjugated) concentration was quantified using the HPLC-MS/MS. We analyzed the association between urinary BPA concentration and the risk of TNs using crude and multivariable logistic regression models. Participants were further stratified into thyroid autoantibody positive group (at least one positive) and thyroid autoantibody negative group (both negative) according to the thyroglobulin antibody (TGAb) and thyroid peroxidase antibody (TPOAb) levels, and restricted cubic spline regression was also applied to determine the possible nonlinear relationship between urinary BPA and TNs.Compared with women in the first quartile, the odds of TNs was 72% (adjusted OR = 1.72, 95% CI: 1.25 to 2.35) higher for those in the second quartile, 54% (adjusted OR = 1.54, 95% CI: 1.12 to 2.12) higher for those in the third quartile, and 108% (adjusted OR = 2.08, 95% CI: 1.50 to 2.90) higher for those in the fourth quartile after adjusting for age, BMI, education, HDL-C, LDL-C, triglyceride, total cholesterol, urinary iodine, TGAb and TPOAb. When the study population was stratified into thyroid autoantibody positive group and thyroid autoantibody negative group, we found that only in the positive group, the association was significant in model 1 (crude OR = 2.80; 95% CI = 1.90 to 4.12), model 2 (adjusted OR = 2.84; 95% CI = 1.91 to 4.22), model 3 (adjusted OR = 4.01; 95% CI = 2.57 to 6.27) and model 4 (adjusted OR = 3.71; 95% CI = 2.36 to 5.83). Multivariable-adjusted restricted cubic spline analysis demonstrated a similar result that in the thyroid autoantibody positive group, the association between urinary BPA and TNs risk was near linear (P-overall <0.001; P-non-linear = 0.054).In Chinese women, higher urinary BPA concentration was associated with increased risk of TNs only in those with positive thyroid autoantibodies. Moreover, this association was near linear, indicating that any rise in BPA exposure was associated with elevated TNs risk.
