P1536Does free thyroxine (fT4) predict risk of atherosclerotic cardiovascular disease (ASCVD)?

Abstract

Abstract Background Thyroid hormones are associated with arrhythmic risk, but their relationship to ASCVD is unclear. The Rotterdam Study* reported associations of higher fT4 and lower thyroid stimulating hormone (TSH) levels with ASCVD, including within the euthyroid range. Free T3 (fT3) was not assessed. Methods We tested whether fT4, fT3, and TSH levels were associated with ASCVD in Intermountain Healthcare. All patients >18 y old with an fT4 level in the electronic medical record database were included. The hormone reference ranges were divided into quartiles (Q), and associations with prevalent and incident ASCVD were assessed by multivariable regression and trend tests. Results A total of 212,202 patients (age = 64.4±11.2 y; 66.6% women) were included and followed for 6.1±4.4 y. Of these, 8.3%, 86.6%, and 5.1% had fT4 levels below, within, and above the reference range. CAD was prevalent in 18.9% of fT4 Q1 patients. The adjusted odds ratio (OR) for coronary artery disease (CAD) increased through (Q4/Q1 OR=1.36) and beyond (High/Q1 OR=1.71) the normal range, p-trend<0.001 (Table). Smaller incremental risks were noted for fT3 (Q4/Q1 OR=1.13; High/Q1 OR=1.25). The frequencies of incident MI (Q1=2.5%) and stroke (Q1=5.4%) were low and did not show a concentration-related risk gradient. Incident all-cause death (Q1=24.3%) increased slightly for Q4 and high fT4 (OR=1.05, 1.06) but not fT3; death also increased with low fT4 (OR=1.28). TSH showed no consistent gradient within the normal range for prevalent or incident events; however, mortality increased with both high and low TSH. Thyroid Hormone Levels and Prevalent CAD Prevalent CAD (adjusted OR) Low Normal Q1 Normal Q2 Normal Q3 Normal Q4 High fT4 (n=212,202) OR=1.08, p=0.02 OR=1.00 referent OR=1.11, p<0.001 OR=1.21, p<0.001 OR=1.36, p<0.001 OR=1.71, p<0.001 fT3 (n=30,200) OR=1.12, p<0.001 OR=1.00 referent OR=0.98, p=0.61 OR=1.02, p=0.53 OR=1.13, p<0.001 OR=1.25, p<0.001 TSH (n=183,227) OR=1.39, p<0.001 OR=1.00 referent OR=0.93, p=0.42 OR=0.88, p=0.15 OR=0.92, p=0.35 OR=1.73, p<0.001 Reference Q 1–4 are: fT4: 0.75–0.90; 0.91–1.01; 2.02–1.14; 1.15–1.50 ng/dL; TSH: 0.54–1.30; 1.31–2.04; 2.05–3.68; 3.69–6.80 uIU/mL; fT3: 2.40–2.60; 2.70–2.80; 2.90–3.10; 3.20–4.20 pg/dL. Conclusions Consistent with the Rotterdam Study, we found an increase in prevalent CAD with increasing fT4 levels within and beyond the normal range and, uniquely, a more modest relationship with fT3. We could not confirm a normal-range relationship between hormone levels and incident events or between TSH and prevalent disease. The relationship of fT4 levels to ASCVD is intriguing, is deserving of further study, and may have important implications for ASCVD prevention. *A Bano, et-al. Circ Res 2017; 121:1397–1400