Abstract
BACKGROUND: Poor olfaction may be associated with incident heart failure (HF) in older adults, but empirical evidence is scant. METHODS: We included 5217 participants free of clinical HF and with a smell assessment in 2011-2013 from the Atherosclerosis Risk in Communities Study. Olfaction was measured by the 12-item Sniffin' Sticks odor identification test and defined as good (score 11-12), moderate (9-10), or poor (≤8). Participants were followed until the first HF hospitalization, death, last contact, or December 31, 2020, whichever happened first. We estimated adjusted risk ratios (aRR) for associations of olfaction with incident HF and its subtypes, and cross-sectional associations of olfaction with subclinical HF markers, including N-terminal pro-B-type natriuretic peptides (NT-proBNP), high-sensitive cardiac troponin T (hs-cTnT), and echocardiogram-defined structural heart disease. RESULTS: During a median 8.4-year follow-up, we identified 622 incident HF, including 212 with reduced ejection fraction (HFrEF) and 250 with preserved EF (HFpEF). Comparing poor with good olfaction, the aRR of HF was 1.24 (95% confidence interval (CI): 1.03,1.51) at year 8. Moderate olfaction showed a similar association pattern with HF risk, with the corresponding aRR of 1.23 (95% CI, 1.00-1.50). Poor olfaction appeared to have an evident association with HFrEF but not with HFpEF. Poor olfaction was associated with higher median levels of NT-proBNP and hs-cTnT, and higher odds of having structural heart disease than good olfaction. CONCLUSIONS: In older adults, poor olfaction identified by a single smell test was associated with a modestly higher risk of HF, especially HFrEF, and with known subclinical HF biomarkers.