Causal Association of Dietary Habits With Thyroid Diseases: Univariate and Multivariate Mendelian Randomization Studies

Dietary habits play a crucial role in everyday health, influencing both the onset and progression of diseases. Previous studies have shown some effects of dietary habits on certain thyroid diseases. However, comprehensive research on the causal relationship between dietary habits and thyroid diseases is lacking. The purpose of this study was to investigate the potential causal relationship between 45 genetically predicted dietary intake habits and thyroid diseases. We obtained GWAS data for 45 dietary intake habits from the UK Biobank and for 10 thyroid diseases from the FinnGen R12 database. We performed univariable Mendelian randomization (UVMR) using the inverse variance weighted (IVW) test as the primary test and corrected the results by false discovery rate (FDR) analysis. Additionally, multivariable Mendelian randomization analysis (MVMR) was further applied to assess independent effects of dietary habits. Linkage disequilibrium score regression (LDSC) was conducted on the FDR-corrected significant findings. UVMR identified 28 potential causal associations, of which six remained significant after FDR correction. Cheese intake and alcohol consumption during meals were suggestively associated with a lower risk of hypothyroidism. Poultry consumption showed a potential association with a higher risk of nontoxic thyroid nodules, while moderate red wine intake appeared to be linked with a lower risk. In MVMR analyses adjusting for conventional confounders, the association between poultry consumption and both nontoxic nodules and goiter remained suggestive of an independent effect. These findings suggest potential causal links between dietary factors and thyroid diseases; however, considering the MR assumptions and horizontal pleiotropy, these estimates should be interpreted with caution, as they are hypothesis-generating rather than prescriptive for clinical nutrition.