Causal effect of thyroid function on cortical brain structure: A two-sample Mendelian randomization study.

BACKGROUND

The causal relationship between variations in thyroid dysfunction and brain structure remains elusive. We intended to probe into this causal connection using a Mendelian randomization (MR) approach.

METHODS

Aggregated statistics from the genome-wide Association Study (GWAS) on Autoimmune thyroid dysfunction were obtained from the FinnGen database, including autoimmune hyperthyroidism (N = 2,190) and autoimmune hypothyroidism (N = 49,055). Aggregate statistics on cortical area and thickness for GWAS were obtained from the Enhancing NeuroImaging Genetics through Meta-Analysis (ENIGMA) Consortium (N = 51,665). The main method employed is inverse variance weighting (IVW), supplemented by weighted median, Mr-Egger regression, MR-pleotropy residuals, and outliers. Some sensitivity analyses were conducted to determine heterogeneity and pleiotropy.

RESULTS

For autoimmune hypothyroidism (exposure), significant causal associations were observed with cortical thickness (outcome type) in the following regions: pars triangularis (β = -0.003, 95% CI: -0.005 to -0.001, PTH = 0.023; PTH: P-value for cortical thickness,), posterior cingulate (β = 0.003, 95% CI: 0.000-0.006, PTH = 0.034), and transverse temporal (β = 0.006, 95% CI: 0.002-0.010, PTH = 0.004; β = 0.006, 95% CI: 0.001-0.011, PTH = 0.010). For autoimmune hyperthyroidism (exposure), significant causal associations were observed with cortical surface area (outcome type) in the entorhinal (β = -2.576, 95% CI: -4.714 to -0.437, PSA = 0.018; PSA: P-value for cortical surface area), frontal pole (β = -1.195, 95% CI: -2.118 to -0.271, PSA = 0.011), lateral occipital (β = -15.787, 95% CI: -31.171 to -0.403, PSA = 0.044), lingual (β = -10.778, 95% CI: -20.888 to -0.668, PSA = 0.037), postcentral (β = -16.360, 95% CI: -28.410 to -4.309, PSA = 0.008), precentral (β = -16.089, 95% CI: -29.281 to -2.897, PSA = 0.017), and superior parietal (with global weighting; β = 11.360, 95% CI: 0.435-22.285, PSA = 0.042). Additionally, autoimmune hyperthyroidism was causally associated with cortical thickness (outcome type) in the entorhinal (β = 0.009, 95% CI: 0.000-0.018, PTH = 0.039) and posterior cingulate (β = 0.004, 95% CI: 0.001-0.008, PTH = 0.025). These findings remained reliable in the sensitivity analyses.

CONCLUSIONS

These findings suggest that autoimmune hyperthyroidism and autoimmune hypothyroidism may have potential causal effects on the surface area and thickness of specific regions of the cerebral cortex, but further studies are needed to fully understand the nature of this relationship.