Thyroid Stimulating Hormone Levels Are Associated With Genetically Predicted Nonalcoholic Fatty Liver Disease

Context Findings from observational studies indicate an association of thyroid hormone levels with the risk of nonalcoholic fatty liver disease (NAFLD); however, conflicting results remain and reverse causality may be a possibility. Objective This study aimed to evaluate the associations between NAFLD and both plasma thyroxine (T4) and thyroid stimulating hormone (TSH) at the phenotypic and genetic levels. Methods We included 14 797 participants, aged 20 to 74 years who had undergone abdominal ultrasonography during the Third National Health and Nutrition Examination Survey (NHANES III). Multivariable logistic regression analyses were used to examine the observational associations of TSH and T4 with NAFLD. Mediation analyses were performed to study whether the relationship between NAFLD and TSH levels was mediated via potential confounders. A bidirectional, two-sample Mendelian randomization (MR) analysis was used to determine the potential causal relationship. Results Multivariable logistic regression model suggested a "dose-response" relationship between TSH (Q4 vs Q1: OR = 1.29; 95% CI, 1.10-1.52; P-trend = 0.001) and NAFLD. BMI and ALT partially mediated the association between TSH and NAFLD, while the proportion of the mediation effects of BMI and ALT were 39.1% and 22.3%, respectively. In MR analyses, the inverse-variance weighted method was selected as primary method and suggested a putative causal effect of NAFLD on serum TSH levels (OR = 1.022; 95% CI, 1.002-1.043). The result was further validated in the sensitivity analyses. Conclusion Circulating TSH levels were associated with the risk of NAFLD. MR analysis suggested a putative causal effect of NAFLD on TSH levels.