LAUSR

Abnormal thyroid hormone concentrations have been associated with adverse cardiovascular outcomes, but the relationship between thyroid dysfunction and specific heart failure phenotypes is less clear. To examine the association of thyroid dysfunction with the risk of incident HF in older adults without pre-existing HF. We analyzed participants enrolled in the Atherosclerosis Risk in Communities (ARIC) study who attended the visit 5 examination (2011–2013). Participants with previous HF history, and participants treated with amiodarone, levothyroxine, and antithyroid medication were excluded. We used Cox regression models to assess the associations between serum thyroid indices (free thyroxine [FT4], total triiodothyronine [TT3], or thyroid stimulating hormone [TSH]) and incident adjudicated HF with reduced (HFrEF) and preserved (HFpEF) left ventricular ejection fraction. Continuous associations between TT3 and outcome were further assessed via Cox model using restricted cubic spline. Among 3349 participants (mean age 75±5 years, 56% women, 20% black), subclinical hypothyroidism was prevalent in 12% of participants and low T3 syndrome in 3%. Those with overt hypothyroidism (<1%) or hyperthyroidism (<1%) were not included in the analysis given the low prevalence. Over a median follow-up of 5.5 years, incident HF occurred in 198 subjects (5.9%) at a rate of 11.1 per 1000 person-years. Of these, 86 were HFrEF, 83 HFpEF, and 29 were unclassified HF. We observed an inverse association of TT3 level with risk of incident HFpEF, but not overall incident HF or incident HFrEF, after adjustment for clinical confounders and baseline NT-proBNP levels (HR per 1 SD 0.70, 95% CI 0.54–0.92; P 0.010) (Figure). Similar results were observed for the composite endpoint of incident HFpEF or all-cause death. No statistically significant associations were found between TSH or T4 levels and incident HF. Low T3 syndrome was associated with incident HFpEF, but not overall incident HF or incident HFrEF, after adjustment for clinical confounders (HR 2.71, 95% CI 1.08–6.82; P 0.035); however, its association was significantly attenuated after adjustment with NT-proBNP (HR 2.25, 95% CI 0.87–5.79; P 0.09). No statistically significant association was found between subclinical hypothyroidism and incident HF. In a contemporary biracial cohort of older adults, serum T3 level was inversely associated with incident HFpEF hospitalization. T3 administration could be considered as a potential target in future clinical trials preventing HFpEF hospitalization. Figure 1 Type of funding source: Public Institution(s). Main funding source(s): The Atherosclerosis Risk in Communities Study is performed as a collaborative study supported by National Heart, Lung, and Blood Institute contracts