Higher thyrotropin leads to unfavorable lipid profile and somewhat higher cardiovascular disease risk: evidence from multi-cohort Mendelian randomization and metabolomic profiling.
van Vliet NA., Bos MM., Thesing CS., Chaker L., Pietzner M., Houtman E., Neville MJ., Li-Gao R., Trompet S., Mustafa R., Ahmadizar F., Beekman M., Bot M., Budde K., Christodoulides C., Dehghan A., Delles C., Elliott P., Evangelou M., Gao H., Ghanbari M., van Herwaarden AE., Ikram MA., Jaeger M., Jukema JW., Karaman I., Karpe F., Kloppenburg M., Meessen JMTA., Meulenbelt I., Milaneschi Y., Mooijaart SP., Mook-Kanamori DO., Netea MG., Netea-Maier RT., Peeters RP., Penninx BWJH., Sattar N., Slagboom PE., Suchiman HED., Völzke H., Willems van Dijk K., Noordam R., van Heemst D., BBMRI Metabolomics Consortium None.
BACKGROUND: Observational studies suggest interconnections between thyroid status, metabolism, and risk of coronary artery disease (CAD), but causality remains to be proven. The present study aimed to investigate the potential causal relationship between thyroid status and cardiovascular disease and to characterize the metabolomic profile associated with thyroid status. METHODS: Multi-cohort two-sample Mendelian randomization (MR) was performed utilizing genome-wide significant variants as instruments for standardized thyrotropin (TSH) and free thyroxine (fT4) within the reference range. Associations between TSH and fT4 and metabolic profile were investigated in a two-stage manner: associations between TSH and fT4 and the full panel of 161 metabolomic markers were first assessed hypothesis-free, then directional consistency was assessed through Mendelian randomization, another metabolic profile platform, and in individuals with biochemically defined thyroid dysfunction. RESULTS: Circulating TSH was associated with 52/161 metabolomic markers, and fT4 levels were associated with 21/161 metabolomic markers among 9432 euthyroid individuals (median age varied from 23.0 to 75.4 years, 54.5% women). Positive associations between circulating TSH levels and concentrations of very low-density lipoprotein subclasses and components, triglycerides, and triglyceride content of lipoproteins were directionally consistent across the multivariable regression, MR, metabolomic platforms, and for individuals with hypo- and hyperthyroidism. Associations with fT4 levels inversely reflected those observed with TSH. Among 91,810 CAD cases and 656,091 controls of European ancestry, per 1-SD increase of genetically determined TSH concentration risk of CAD increased slightly, but not significantly, with an OR of 1.03 (95% CI 0.99-1.07; p value 0.16), whereas higher genetically determined fT4 levels were not associated with CAD risk (OR 1.00 per SD increase of fT4; 95% CI 0.96-1.04; p value 0.59). CONCLUSIONS: Lower thyroid status leads to an unfavorable lipid profile and a somewhat increased cardiovascular disease risk.