Search

Background GDF15 (growth differentiation factor 15) is a potent predictor of bleeding in people with cardiovascular disease. However, whether GDF15 is associated with bleeding in individuals without a history of cardiovascular disease is unknown. Methods and Results The study population was from the ARIC (Atherosclerosis Risk in Communities) study. We studied the association of GDF15 with hospitalized bleeding events among 9205 participants (1993–1995) without prior bleeding and cardiovascular disease (mean age 60 years, 57% women, 21% Black). Plasma levels of GDF15 were measured in relative fluorescence units using DNA‐based aptamer technology. Bleeding was ascertained using discharge codes. We examined hazard ratios (HRs) of incident bleeding using Cox models and risk prediction with the addition of GDF15 to clinical predictors of bleeding. There were 1328 hospitalizations with bleeding during a median follow‐up of 22.5 years. The majority (76.5%) were because of gastrointestinal bleeding. The absolute incidence rate of bleeding per 1000 person‐years was 11.64 in the highest quartile of GDF15 versus 5.22 in the lowest quartile. The highest versus lowest quartile of GDF15 demonstrated an adjusted HR of 2.00 (95% CI, 1.69–2.35) for total bleeding. The findings were consistent when we examined bleeding as the primary discharge diagnosis. The addition of GDF15 to clinical predictors of bleeding improved the C‐statistic by 0.006 (0.002–0.011) from 0.684 to 0.690, P =0.008. Conclusions Higher levels of GDF15 were associated with bleeding events and improved the risk prediction beyond clinical predictors in individuals without cardiovascular disease.

Introduction: Associations of nonalcoholic fatty liver disease (NAFLD) with dementia are controversial. The full spectrum of NAFLD from simple steatosis to fibrosis has been less investigated. Hypothesis: The severity of NAFLD, especially with the stage of liver fibrosis, is associated with dementia. Methods: The associations between midlife (1990-92, N=9283, mean age 57, female 55%) and late-life (2011-13, N=5138, mean age 75, female 58%) NAFLD and all cause dementia were quantified by Cox regression. NAFLD was characterized using the fatty liver index and liver fibrosis assessment model (FIB-4). Dementia was adjudicated and ascertained through Dec. 31, 2019 from neuropsychological assessments, annual participant or informant contact, and medical record surveillance. We used Cox models with adjustment for demographics, APOE ε4, alcohol, and kidney function. Additional adjustments were made for metabolic factors to explore the independent contribution of NAFLD to dementia. Results: During a median follow-up of 24.5 years after midlife NAFLD assessment, 1854 participants developed dementia. During a median of 6.3 years in late-life, there were 893 dementia cases. A graded increase in dementia risk across the spectrum of NAFLD was observed at midlife (p-trend Figure ). In fact, NAFLD in late-life tended to be protective against dementia. After adjusting for the metabolic factors, the associations remained statistically significant in some categories. Conclusions: Remodeling of hepatic architecture and dysregulation in hepatocellular function in NAFLD at midlife plausibly induce peripheral promoters of neurodegeneration and may expose a large segment of the population to increased risk of dementia. The inverse association of NAFLD with dementia at late-life requires further investigation, possibly reflecting the depletion of susceptibles, the reversibility of NAFLD due to weight loss, and the lack of accuracy in identifying NAFLD using clinical prediction models at older age.

Introduction: Eicosanoids play an important role in regulating inflammation response, which influences the development of heart failure (HF). However, the association between eicosanoids and the risk of HF remains unclear. Hypothesis: We hypothesize that eicosanoid metabolites are associated with incident HF and may improve HF prediction. Methods: We measured plasma eicosanoids via liquid chromatography-mass spectrometry among HF-free participants at visit 2 (1990-1992) from the Atherosclerosis Risk in Communities (ARIC) study. The Cox proportional hazard regression was performed to evaluate the relationship between eicosanoids and incident HF controlling for clinical risk factors. The least absolute shrinkage and selection operator was applied to identify a representative subset of eicosanoids towards constructing an eicosanoid risk score (ERS). We further conducted genome-wide association analyses to examine the genetic determinants for HF related eicosanoids in black and white participants respectively. Results: During an average of 21 years of follow up, 2,202 (28.0 % blacks, 50.3% women) out of 9,519 participants developed HF. Out of 200 eicosanoids analyzed, six with hazard ratio between 0.89 to 1.10 were found associated with the risk of HF (FDR UGT2B7 , where its T allele was associated with lower EPA [M-H+Acetate] level while positively affect the risk of HF and systolic blood pressure level. ERS constructed from 52 eicosanoids showed a 39% increase in HF risk per SD increment and twice the risk between the highest and the lowest ERS quartile (p < 0.001). The addition of ERS to the HF 10-years prediction model modestly improved the C-statistics from 0.799 to 0.808, 95%CI ΔC : 0.004-0.015. Conclusion: We identified a set of eicosanoids that significantly associated with increased risk of HF. The contributing genetic variation of eicosanoids may depict the biological pathways related to HF.

Background: Inflammation plays an important role in the pathophysiology of atherosclerosis. However, inflammatory biomarkers have been found to have a weak association with coronary artery calcium (CAC), a representative measure of subclinical atherosclerosis. Moreover, the existing evidence does not sufficiently explore extra coronary calcium (ECC) in this context. Aim: To characterize the association of two inflammatory markers, high-sensitivity C-reactive protein (hs-CRP) and galectin-3, with CAC and ECC in a community-based cohort. Methods: Cardiac CT was performed among 1,934 ARIC participants (age 53-74 years) without coronary heart disease at visit 7 (2018-19). We examined the associations of hs-CRP and galectin-3 measured in middle age at visit 4 (1996-98) by quartile with the presence of CAC and ECC (Agatston score >0 vs. 0) using multivariable logistic regression. Results: Higher hs-CRP was associated with calcifications in the ascending aorta, aortic valve ring, mitral valve and right coronary artery in the unadjusted model (Table). These associations were mostly attenuated after adjusting for potential confounders, but the associations with ascending aorta and right coronary artery calcifications remained significant (adjusted odds ratio (aOR) = 1.45 [95% CI = 1.02-2.07]) and (1.55 [1.12-2.16]) for the highest vs. lowest quartiles, respectively). Galectin-3 was also independently associated with right coronary artery calcification after adjusting for potential confounders (aOR 1.48 [1.02 - 2.01]) for the highest vs. lowest quartiles). These associations were generally consistent in demographic subgroups. Conclusions: Both hs-CRP and galectin-3 were associated with calcification of some but not all vascular beds tested, suggesting potentially unique atherosclerotic pathophysiology across different vascular beds. Robust associations of inflammatory markers with right coronary artery calcification deserve further investigation.

Introduction: Measures of cardiac structure and function provide important diagnostic and prognostic information for heart failure (HF). Few studies have assessed the associations of circulating metabolites with cardiac structure and function. Hypothesis: We hypothesize that circulating metabolites that reflect aging process are associated with cardiac structure and function measures and incident HF. Methods: Participants from the Atherosclerosis Risk in Communities (ARIC) study visit 5 and the Hispanic Community Health Study / Study of Latinos (HCHS/SOL) study visit 1 who had serum metabolite measures but not prevalent HF were included. Linear regressions were used to examine the associations of metabolites with ten cardiac structure and function variables adjusting for clinical risk factors in each race and study strata, followed by random-effect meta-analyses. The Cox regression was applied to examine the relationship between those identified metabolites and incident HF post visit 5 in ARIC. Results: Among 589 analyzed metabolites, 179 were associated with cardiac structure or function measures in 706 Blacks, 3,358 Whites, and 1,380 Hispanics (FDR < 0.05). Forty-one metabolites were related to two or more measures, where 22 were associated with incident HF (308 HF cases, p Conclusions: We identified multiple metabolites associated with cardiac structure and function measures in multi-ethnic populations, highlighting metabolic pathways in aging and their impact on HF.

Introduction: Proteins are involved in the pathophysiology of hypertension (HTN), but there are no large scale prospective proteomic analyses of incident HTN. Hypothesis: To identify and validate plasma proteins associated with incident HTN among a bi-racial cohort in the US. Methods: We quantified the associations of 4,955 plasma proteins (measured by SomaScan version 4 assay) with incident HTN over median follow-up of 11 years among 5,080 participants without baseline hypertension at ARIC visit 3 (1993-95). We then validated significant proteins from the primary analysis with incident HTN between visit 2 (n=6,810 during 1990-92) and visit 3 (median follow-up of 3.1y), to avoid including the same incident HTN cases in both analyses. Incident HTN was defined as systolic blood pressure (BP) ≥140, diastolic BP ≥90 mmHg, or HTN medication use. We ran multivariable Cox models and applied Benjamin-Hochberg procedure, with false discovery rate of 0.05, to account for multiple comparisons. Results: In the primary analysis with visit 3 as baseline (mean age 59y, 54% women, 13% Black), 3,828 participants (75%) developed HTN, and we found 55 proteins significantly associated with incident HTN. Among them, 14 proteins were validated using visit 2 data (Figure). Of those, 11 showed positive associations with HTN, and they are involved in glycoprotein degradation (e.g., MMP7, HTRA1, ASGR1, GUSB, CPM), cell adhesion (e.g., LGALS3BP and RET), and inflammation (e.g., CRP). In contrast, 3 proteins, renin ([REN]), netrin receptor ([UNC5D] apoptosis) and secretogranin-3([SCG3] endothelial cell growth), had inverse relationships with incident HTN. Further exploration showed a U-shaped association between renin and incident HTN. Conclusion: We identified several proteins associated with incident HTN, with plausible functions contributing to the development of HTN. These findings have implications on understanding mechanisms of HTN and potential targets for prevention and treatment.

Background: Physical activity (PA) patterns over time influence diabetes risk through incompletely understood mechanisms. Adipokines secreted by adipose tissue have myriad endocrine effects and are affected by PA. It is unknown the extent to which adipokines mediate the association between changes in PA and diabetes. Methods: We conducted a cross-sectional study of 10,908 ARIC participants (mean age 54; 55% female; 21% Black adults) who attended Visits 1 (V1; 1987-1989) and 3 (V3; 1993-1995) of the ARIC Study. We used mediation analyses to determine the extent to which the association of changes in PA from V1 to V3 and diabetes status at V3 was explained by adipokines. Analyses were stratified by obesity status. PA was defined using American Heart Association guidelines as inactive, intermediate, and recommended (150 minutes of moderate-vigorous activity/week); the intermediate group was further divided into low and moderate (below or above the median activity level). Participants were categorized by PA at V1 and V3 - (1) stable low: inactive or low at both visits; (2) increasing: higher category at V3 than V1; (3) decreasing: lower category at V3 than V1; and (4) stable high: moderate or high at both visits. Results: Increasing PA was associated with higher adiponectin and lower leptin at V3 compared to stable low PA, while decreasing PA was associated with lower adiponectin and higher leptin compared to stable high PA (p Conclusions: The associations between changes in PA and diabetes may be mediated in part by adipokines.

Summary: Background: Galectin-3 (gal-3) is a β-galactoside-binding lectin associated tissue fibrosis and inflammation. There is limited understanding of the relationship between gal-3 and vascular health. Our aim was to assess the association between gal-3 and arterial stiffness in older adults. Methods: We conducted a cross-sectional study of 4275 participants (mean age of 75 years) from the Atherosclerosis Risk in Communities (ARIC) Study. Central arterial stiffness was measured by carotid-femoral pulse wave velocity (cfPWV). We evaluated the association of gal-3 with cfPWV using multivariable linear regression. Results: The median (interquartile range) gal-3 concentration was 16.5 (13.8, 19.8) ng/mL and mean cfPWV was 1163±303 cm/s. Higher gal-3 concentration was associated with greater central arterial stiffness after adjustment for age, sex, race-center, heart rate, systolic blood pressure, anti-hypertensive medication use, and current smoking status (β=36.4 cm/s change in cfPWV per log unit change in gal-3; 95% CI: 7.2, 65.5, p=0.015). The association was attenuated after adjusting for additional cardiovascular risk factors (β=17.3, 95% CI: −14.4, 49.0). Conclusions: In community-dwelling older adults, gal-3 concentration was associated with central arterial stiffness, likely sharing common pathways with traditional cardiovascular risk factors.

Background Plasma lipoprotein(a) (Lp[a]) concentrations are primarily determined by genetic factors and are believed to remain stable throughout life. However, data are scarce on longitudinal trends in Lp(a) concentrations over time. Therefore, it is unclear whether measurement of Lp(a) once in a person's life is sufficient for cardiovascular risk assessment in all adults. Methods and Results Lp(a) concentrations, specifically apolipoprotein(a) concentrations, were measured at visits 4 and 5, ≈15 years apart, in 4734 adult participants of the ARIC (Atherosclerosis Risk in Communities) study (mean age at visits 4 and 5, 60.7±5.1 and 75.5±5.2 years, respectively). Participants were categorized by baseline (visit 4) Lp(a) concentrations as normal ( Conclusions Our results suggest that adults with borderline‐high Lp(a) concentrations may be considered for repeat monitoring of Lp(a) over time, particularly if they are Black, women, or have diabetes, hypertension, and/or elevated albuminuria.

Background Circulating high sensitivity cardiac troponin T (hs‐cTnT) is associated with incidence of atrial fibrillation (AF), but the association of changes in hs‐cTnT over time on incident AF has not been explored. Hypothesis Six‐year increase in circulating hs‐cTnT will be associated with increased risk of AF and will contribute to improved prediction of incident AF. Methods We conducted a prospective cohort analysis of 8431 participants from the Atherosclerosis Risk in Communities (ARIC) study. hs‐cTnT change was categorized at visit 2 and 4 as undetectable (

Background Although intensive blood pressure reduction has cardiovascular benefits, the absolute benefit is greater in those at higher cardiovascular disease (CVD) risk. Objectives This study examined whether N-terminal pro-B-type natriuretic peptide (NT-proBNP) helps identify subjects at higher risk for CVD events across systolic blood pressure (SBP), diastolic blood pressure (DBP), or pulse pressure (PP) categories. Methods Participants from the ARIC (Atherosclerosis Risk In Communities) study visit 4 (1996 to 98) were grouped according to SBP, DBP, or PP categories and further stratified by NT-proBNP categories. Cox regression models were used to estimate hazard ratios for incident CVD (coronary heart disease, ischemic stroke, or heart failure hospitalization) and mortality across combined NT-proBNP and/or BP categories, adjusting for CVD risk factors. Results There were 9,309 participants (age: 62.6 ± 5.6 years; 58.3% women) with 2,416 CVD events over a median follow-up of 16.7 years. Within each SBP, DBP, or PP category, a higher category of NT-proBNP (100 to Conclusions Elevated NT-proBNP is independently associated with CVD and mortality across SBP, DBP, and PP categories and helps identify subjects at the highest risk. Participants with stage 1 hypertension but elevated NT-proBNP had greater cardiovascular risk compared with those with stage 2 SBP but lower NT-proBNP. Future studies are needed to evaluate use of biomarker-based strategies for CVD risk assessment to assist with initiation or intensification of BP treatment.

Background Lipoprotein(a) (Lp(a)) is a potent causal risk factor for cardiovascular events and mortality. However, its relationship with subclinical atherosclerosis, as defined by arterial calcification, remains unclear. This study uses the ARIC (Atherosclerosis Risk in Communities Study) to evaluate the relationship between Lp(a) in middle age and measures of vascular and valvular calcification in older age. Methods and Results Lp(a) was measured at ARIC visit 4 (1996–1998), and coronary artery calcium (CAC), together with extracoronary calcification (including aortic valve calcium, aortic valve ring calcium, mitral valve calcification, and thoracic aortic calcification), was measured at visit 7 (2018–2019). Lp(a) was defined as elevated if >50 mg/dL and CAC/extracoronary calcification were defined as elevated if >100. Logistic and linear regression models were used to evaluate the association between Lp(a) and CAC/extracoronary calcification, with further stratification by race. The mean age of participants at visit 4 was 59.2 (SD 4.3) years, with 62.2% women. In multivariable adjusted analyses, elevated Lp(a) was associated with higher odds of elevated aortic valve calcium (adjusted odds ratio [aOR], 1.82; 95% CI, 1.34–2.47), CAC (aOR, 1.40; 95% CI, 1.08–1.81), aortic valve ring calcium (aOR, 1.36; 95% CI, 1.07–1.73), mitral valve calcification (aOR, 1.37; 95% CI, 1.06–1.78), and thoracic aortic calcification (aOR, 1.36; 95% CI, 1.05–1.77). Similar results were obtained when Lp(a) and CAC/extracoronary calcification were examined on continuous logarithmic scales. There was no significant difference in the association between Lp(a) and each measure of calcification by race or sex. Conclusions Elevated Lp(a) at middle age is significantly associated with vascular and valvular calcification in older age, represented by elevated CAC, aortic valve calcium, aortic valve ring calcium, mitral valve calcification, thoracic aortic calcification. Our findings encourage assessing Lp(a) levels in individuals with increased cardiovascular disease risk, with subsequent comprehensive vascular and valvular assessment where elevated.

Background Laboratory data suggest obesity is linked to myocardial inflammation and fibrosis, but clinical data are limited. We aimed to examine the association of obesity with galectin‐3, a biomarker of cardiac inflammation and fibrosis, and the related implications for heart failure (HF) risk. Methods and Results We evaluated 8687 participants (mean age 63 years; 21% Black) at ARIC (Atherosclerosis Risk in Communities) Visit 4 (1996–1998) who were free of heart disease. We used adjusted logistic regression to estimate the association of body mass index (BMI) categories with elevated galectin‐3 (≥75th sex‐specific percentile) overall and across demographic subgroups, with tests for interaction. We used Cox proportional hazards models to assess the combined associations of galectin‐3 and BMI with incident HF (through December 31, 2019). Higher BMI was associated with higher odds of elevated galectin‐3 (odds ratio [OR], 2.32; 95% CI, 1.88–2.86) for severe obesity ([BMI ≥35 kg/m 2 ] versus normal weight [BMI 18.5‐2 ]). There were stronger associations of BMI with elevated galectin‐3 among women versus men and White versus Black participants (both P ‐for‐interaction 4‐fold higher risk of HF (HR, 4.19; 95% CI, 2.98–5.88) than those with normal weight and galectin‐3 Conclusions Obesity is strongly associated with elevated galectin‐3. Additionally, the combination of obesity and elevated galectin‐3 is associated with marked HF risk, underscoring the importance of elucidating pathways linking obesity with cardiac inflammation and fibrosis.

Introduction: Animal studies and emerging population data link liver function and integrity to brain health, although mostly based on cross-sectional estimates. We reported that low liver transaminases are associated with risk of dementia, but prospective studies on liver biomarkers and altered brain morphology are lacking. Methods: We studied 1,596 participants of the Atherosclerosis Risk in Communities (ARIC) Study cohort examined in 1996-98 (mean age: 62 years, 59% female, 27% Black) free of chronic liver disease, cirrhosis, and stroke, and with alanine aminotransferase (ALT) to aspartate aminotransferase (AST) ratio ≤2. Brain magnetic resonance imaging (MRI) scans were performed in 2011-13, at the time of re-examination. Serum transaminases (ALT, AST) and gamma-glutamyl transferase (GGT) at baseline were analyzed as sex-specific quintiles, with the first quintile further subdivided into a th and 10 th -20 th percentiles to allow for nonlinear associations at low biomarker levels. Brain MRI markers were measured with 3 Tesla MRI. Multinomial logistic and linear regressions were used, adjusted for demographics and APOE-ε4. Results: Monotonic associations were observed between higher GGT levels and smaller cortical volume and temporal lobe volume meta regions of interest (Figure) . The lowest GGT levels were associated with smaller volumes of white matter hyperintensities. In contrast, associations of transaminases with brain morphological measures were inconclusive. Conclusions: GGT levels measured in late mid-life are associated vascular sequelae on brain MRI among older adults. As a potent regulator of antioxidant capacity and a predictor of atherosclerosis, GGT may point to vascular pathways that influence brain health. The lack of association between low transaminases and brain MRI findings suggests non-vascular pathways for their association with incident dementia. A deeper understanding of the links between hepatic metabolic dysregulation and brain health is warranted.

Introduction: Heart Failure (HF) has shared genetic architecture with its risk factors, including atrial fibrillation (AF), body mass index (BMI), coronary heart disease (CHD), systolic blood pressure (SBP), and type 2 diabetes (T2D). The risk prediction performance of polygenic risk scores (PRS) for those HF risk factors and HF itself over an established risk equation warrants investigation. Methods: Within the Atherosclerosis Risk in Communities (ARIC) study, six PRSs were constructed for AF, BMI, CHD, SBP, T2D, and HF by summing the product of pre-computed weights from genome-wide association studies and SNP allele dosages in European and African Americans separately. The association between PRSs and incident HF was assessed using cox proportional hazard models, and the 10-, 20-, and 30-year risk prediction performance of PRS over the ARIC HF risk equation was assessed using C-statistics. Associations between AF PRS and HF subtypes, echocardiographic measures, and 4,877 proteins were examined. Results: Over 30 years follow-up, 1,922 (22%) and 735 (29%) HF cases developed in 8,624 European (mean age=54.2, 52% female) and 2,525 African (mean age=53.3, 61% female) Americans. The PRSs for AF and HF were associated with incident HF in both European and African Americans (P-4 ). Protein analyses revealed that 61 proteins were associated with AF PRS, where NT-proBNP and Coagulation factor X showed the strongest positive and negative associations respectively. Conclusions: The PRS of AF was associated with incident HF, and had significant incremental value over an established HF risk prediction equation. These findings suggest that PRS may be useful in identifying individuals with high risk of HF.

Introduction: Frailty, an increased vulnerability to stressors, is associated with adverse health outcomes. Cardiovascular disease (CVD) is associated with frailty, but whether circulating cardiac biomarkers of subclinical myocardial injury and strain (e.g., high-sensitivity cardiac troponin T [hs-cTnT] and natriuretic peptide [NT-proBNP]) are associated with frailty is unknown. Methods: We conducted a cross-sectional analysis of levels of hs-cTnT and NT-proBNP assessed among participants of the ARIC study at visit 5 (2011-2013, mean age: 75.5 years, 42% male, 21% Black). We used the Fried frailty phenotype that counts the presence of each component (weight loss, low physical activity, slow gait, exhaustion, and low grip strength) to classify participants as frail (≥3), pre-frail (1-2), or robust (0 components). Participants without prevalent CVD (stroke, myocardial infarction, or heart failure) were classified into hs-cTnT and NT-proBNP quartiles; those with prevalent CVD were considered in a separate group. Using adjusted multinomial logistic regression models, we tested the hypothesis that hs-cTnT and NT-proBNP are positively associated with odds of being pre-frail and frail, vs. robust. Results: Among 5,162 participants, 1,039 had prevalent CVD (20%), 2,447 (48%) were pre-frail, and 358 (7%) frail. Biomarker concentrations had graded associations with frailty and pre-frailty, vs. robust. Estimates comparing the 4 th vs. 1 st quartile, (e.g., hs-cTnT frail vs. robust OR=4.37 [95% CI: 2.69-7.09]) were similar to those observed for prevalent CVD vs. 1 st quartile (e.g., hs-cTnT frail vs. robust OR=5.11 [3.19-8.17]). All frailty components were positively associated with cardiac biomarker concentrations and CVD (Table), except exhaustion with NT-proBNP. Conclusions: Cardiac biomarker concentrations, even in the absence of CVD, may help identify people at greater risk for frailty. Future research should examine the role of changes in biomarkers from mid-life on the risk of frailty

Background: PAD is associated with substantial morbidity and mortality, yet it is underdiagnosed mainly because many patients are asymptomatic and its screening is not widely conducted. Proteomic analysis may help identify predictors of PAD. Aims: To identify protein predictors of PAD and examine their predictive performance of incident and prevalent PAD beyond traditional risk factors in the ARIC study. Methods: Of 4,877 plasma proteins measured by SOMAscan, we first selected panels of proteins at ARIC visit 3 (N=10,302 participants free of PAD; 1993-95) associated with incident PAD in two steps: 1) we identified significant proteins using multivariable Cox models accounting for multiple comparisons; and 2) we selected proteins using LASSO Cox regression. We constructed 4 protein panels according to λ in LASSO and parsimony. We then evaluated c-statistics after adding each protein panel to a base model with established CVD risk factors. We finally used cross-sectional data at visit 5 (N=4,525; 2011-13) and logistic models to assess if these panels better identify older adults with prevalent PAD beyond traditional predictors. Incident PAD since visit 3 was based on hospitalization with PAD diagnosis; prevalent PAD at visit 5 was defined by ankle-brachial index ≤0.9. Results: There were 450 incident PAD cases after visit 3 (median follow-up: 23 years) and 344 prevalent PAD cases at visit 5. Of 103 proteins significantly associated with incident PAD, 4 protein panels were selected, including 5, 10, 34, and 56 proteins (e.g., angiopoietin-2 and renin), respectively. All 4 protein panels significantly improved c-statistics for incident PAD (e.g., 0.854 for base plus panel 4 vs. 0.807 for base model) (Figure 1A). The same protein panels improved the c-statistic for prevalent PAD at visit 5 compared to the base model (Figure 1B). Conclusions: Protein panels improved risk discrimination for both incident and prevalent PAD beyond established risk factors, with implications for targeted preventive therapies and screening of PAD.

Introduction: Matrix metalloproteinases (MMPs), belong to a family of zinc-dependent endopeptidases, have been found to induce stroke through its role in extracellular matrix destruction. MMPs also affect the stroke prognosis by disrupting blood brain barrier during stroke. Genetic association studies suggest that MMPs are involved in the pathogenesis of both ischemic and hemorrhagic stroke. Hypothesis: High MMP-2, MMP-9 and ratio of MMP-9/TIMP-1 are associated with higher risk of ischemic and hemorrhagic stroke. Method: After exclusion of missing data and prevalent stroke, 11,393 ARIC participants were followed from midlife (visit 3; 1993-1995) and 4,924 participants followed from late life (visit 5; 2011-2013) to 2019. Plasma levels of MMPs were measured by SOMAscan version 4 assay. Stroke cases were reviewed and adjudicated by neurologists. We used Cox models to examine the relationship between MMPs and stroke risk. Result: Mean age of participants were 60±5.7 at visit 3 and 76±5.2 at visit 5. At visit 3 55% were female and 21% Black. A total of 1,013 ischemic and 127 hemorrhagic strokes occurred over 22.34 years of follow-up for the visit 3 analysis and 319 ischemic and 30 hemorrhagic strokes over 7.24 years for the visit 5 analysis. A significant and positive association was detected for MMP-2 at visit 3 baseline for hemorrhagic stroke (HR (95% CI): 2.28 (1.14-4.57)) and at visit 5 baseline with ischemic stroke (1.65 (1.08-2.53)), after covariate adjustment. No associations were observed with MMP-9 and TIMP-1. Conclusion: Results of MMP-2 indicate a possible relationship with MMPs and stroke subtypes, but were not consistent. Though no significance were found for MMP-9 and TIMP-1, their direction of associations with stroke was consistent with the hypothesis.

Background: Cardiac troponins are the cornerstone for the detection of heart attacks. Beyond the acute setting, there is growing evidence that high-sensitivity cardiac troponin T (hs-cTnT) is associated with long-term cardiovascular disease (CVD) risk. The implications of minute elevations in hs-cTnT on short-term CVD risk in the general population has not been investigated. Methods: We included 12,632 adults (43% men, 25% black, mean age: 57 years) in the Atherosclerosis Risk in Communities (ARIC) study who had at least one measurement of hs-cTnT (Roche) across visits 2 (1990-1992), 3 (1996-1998), and 4 (2011-2013) without diagnosed CVD, defined as coronary heart disease, heart failure, or stroke. We used bootstrapping to generate confidence intervals for the cumulative incidence of CVD and generalized estimating equation log-binomial models to quantify the short-term relative rates of incident CVD. Results: Overall, 3.8% (n=479) had elevated hs-cTnT ≥ 14ng/L at their earliest visit; of whom 65% (n=311) had intermediate or lower predicted 10-year risk by the Pooled Cohort Equation. The cumulative 1-year incidence of CVD among those with hs-cTnT ≥ 14ng/L was 5.7% (95% CI: 4.0%, 7.4%) compared to 1.0% (0.9%, 1.1%) among those with hs-cTnT < 6 ng/L and 1.7% (1.3%, 2.2%) among those with hs-cTnT between 6 to 14 ng/L. Among those with hs-cTnT ≥ 14ng/L, the cumulative incidence of CVD at 2 years was 9.6% (7.3%, 11.8%) and at 3-years was 12.7% (10.0%, 15/3%) (Figure 1A). After accounting for traditional CVD risk factors, hs-cTnT ≥ 14ng/L was associated with 2.9 (2.0, 4.1) times higher incidence of CVD at 1-year compared to 1.9 (1.5, 2.4) times incidence at 3-years (Figure 1B). Conclusions: Middle-aged adults in the general population without CVD who had an elevated hs-cTnT experienced a 5.7% 1-year risk of CVD compared to 1% among those with low hs-cTnT. The relative risk associated with elevations in hs-cTnT was most pronounced within the first year, supporting its use for short-term risk monitoring as part of annual preventive visits.

Background: Osteoprotegerin (OPG), a protein involved in bone metabolism, is associated with cardiac fibrosis and remodeling in animal models. Circulating OPG levels correlate with left ventricular mass index and plasma brain natriuretic peptide levels in humans, but the association with incident heart failure (HF) is not known. Methods: Among 10,261 participants in the community-based Atherosclerosis Risk in Communities (ARIC) Study free of prevalent HF at study Visit 3, we investigated the association of plasma OPG levels at Visit 3 with incident HF hospitalization post-Visit 3 using multivariable Cox proportional hazard models. OPG was measured using a high-throughput DNA aptamer-based proteomic platform (Somalogic, Boulder, CO), and relative fluorescence unit values were log2 transformed. All models adjusted for age, sex, race, study field center, body mass index, hypertension, diabetes, estimated glomerular filtration rate, smoking status, prevalent coronary artery disease, and atrial fibrillation. Results: Mean age was 60±6 years, 54% were women, and 20% of participants were Black. Higher OPG levels were associated with older age, female gender, Black race, and higher prevalence of hypertension, diabetes, and smoking (all p Conclusion: In this large community-based cohort, higher plasma OPG levels were significantly associated with incident HF hospitalization, independent of traditional cardiovascular risk factors and NT-proBNP. Further studies are warranted to clarify whether OPG could be a new therapeutic target for HF prevention.

Objective: A Mediterranean diet supplemented with olive oil and nuts prevents cardiovascular disease in clinical studies, but the underlying mechanisms are incompletely understood. We investigated whether the preventive effect of the diet could be due to inhibition of atherosclerosis and foamy monocyte formation in Ldlr –/– mice fed with a diet in which milkfat in a Western diet (WD) was replaced with extra-virgin olive oil and nuts (EVOND). Approach and Results: Ldlr –/– mice were fed EVOND or a Western diet for 3 (or 6) months. Compared with the Western diet, EVOND decreased triglyceride and cholesterol levels but increased unsaturated fatty acid concentrations in plasma. EVOND also lowered intracellular lipid accumulation in circulating monocytes, indicating less formation of foamy monocytes, compared with the Western diet. In addition, compared with the Western diet, EVOND reduced monocyte expression of inflammatory cytokines, CD36, and CD11c, with decreased monocyte uptake of oxLDL (oxidized LDL [low-density lipoprotein]) ex vivo and reduced CD11c + foamy monocyte firm arrest on vascular cell adhesion molecule-1 and E-selectin–coated slides in an ex vivo shear flow assay. Along with these changes, EVOND compared with the Western diet reduced the number of CD11c + macrophages in atherosclerotic lesions and lowered atherosclerotic lesion area of the whole aorta and aortic sinus. Conclusions: A diet enriched in extra-virgin olive oil and nuts, compared with a Western diet high in saturated fat, lowered plasma cholesterol and triglyceride levels, inhibited foamy monocyte formation, inflammation, and adhesion, and reduced atherosclerosis in Ldlr –/– mice.

Isolated systolic hypertension (ISH) and elevated pulse pressure (PP) are common blood pressure (BP) abnormalities in older adults, reflect poor vascular compliance, and can signify risk for cardiovascular outcomes. We sought to characterize the associations of ISH and widened PP with high-sensitivity Troponin-T (hs-cTnT; a marker of myocardial damage) and N-terminal pro-B-type natriuretic peptide (NT-proBNP; a marker of hemodynamic stress) levels in older adults. We performed a cross-sectional analysis of 5,251 Atherosclerosis Risk in Communities (ARIC) study participants without heart failure who attended visit 5 (2011 to 2013). We used logistic regression to evaluate the association of ISH (systolic BP ≥140 mm Hg and diastolic BP < 90 mm Hg) and quartiles of PP with detectable (≥5 ng/L) and elevated hs-cTnT (≥14 ng/L); as well as elevated NT-proBNP (≥100 pg/mL). The mean age was 75 years, 58% were women, and 78% were white. ISH was present in 24.7% and PP ≥ 70 mm Hg in 30.3% of this cohort. Compared to participants with nonhypertensive BP (

Background: We assessed whether plasma troponin I measured by a high-sensitivity assay (hs-TnI) is associated with incident cardiovascular disease (CVD) and mortality in a community-based sample without prior CVD. Methods: ARIC study (Atherosclerosis Risk in Communities) participants aged 54 to 74 years without baseline CVD were included in this study (n=8121). Cox proportional hazards models were constructed to determine associations between hs-TnI and incident coronary heart disease (CHD; myocardial infarction and fatal CHD), ischemic stroke, atherosclerotic CVD (CHD and stroke), heart failure hospitalization, global CVD (atherosclerotic CVD and heart failure), and all-cause mortality. The comparative association of hs-TnI and high-sensitivity troponin T with incident CVD events was also evaluated. Risk prediction models were constructed to assess prediction improvement when hs-TnI was added to traditional risk factors used in the Pooled Cohort Equation. Results: The median follow-up period was ≈15 years. Detectable hs-TnI levels were observed in 85% of the study population. In adjusted models, in comparison to low hs-TnI (lowest quintile, hs-TnI ≤1.3 ng/L), elevated hs-TnI (highest quintile, hs-TnI ≥3.8 ng/L) was associated with greater incident CHD (hazard ratio [HR], 2.20; 95% CI, 1.64–2.95), ischemic stroke (HR, 2.99; 95% CI, 2.01–4.46), atherosclerotic CVD (HR, 2.36; 95% CI, 1.86–3.00), heart failure hospitalization (HR, 4.20; 95% CI, 3.28–5.37), global CVD (HR, 3.01; 95% CI, 2.50–3.63), and all-cause mortality (HR, 1.83; 95% CI, 1.56–2.14). hs-TnI was observed to have a stronger association with incident global CVD events in white than in black individuals and a stronger association with incident CHD in women than in men. hs-TnI and high-sensitivity troponin T were only modestly correlated ( r =0.47) and were complementary in prediction of incident CVD events, with elevation of both troponins conferring the highest risk in comparison with elevation in either one alone. The addition of hs-TnI to the Pooled Cohort Equation model improved risk prediction for atherosclerotic CVD, heart failure, and global CVD. Conclusions: Elevated hs-TnI is strongly associated with increased global CVD incidence in the general population independent of traditional risk factors. hs-TnI and high-sensitivity troponin T provide complementary rather than redundant information.

Background: High-sensitivity cardiac troponin T (TnT) is a potent predictor of cardiovascular disease (CVD) in the general population. Recently, the US FDA has approved Roche fifth generation (Gen 5) TnT assay (more sensitive than the fourth generation [Gen 4] TnT assay). Since many previous epidemiological studies used Gen 4 TnT, it is important to characterize the association of Gen 5 TnT with major CVD events. Methods: We first assessed correlation of Gen 5 vs. Gen 4 TnT in a subsample of 91 participants. Then, as the main analysis, we examined the association of Gen 5 TnT at visit 3 (1993-1995) with major CVD events (coronary heart disease [CHD], stroke, heart failure [HF], and peripheral artery disease [PAD]). Gen 5 TnT was categorized as Results: Gen 5 TnT and Gen 4 TnT had a correlation coefficient of 0.98 (0.88 after excluding outliers >3SD). Of 11,979 participants (mean age 60 [SD 6] years, 1,840 [15%] with prevalent CVD), 5,856 (49%) participants had quantifiable levels of TnT. During a median follow-up of 22.1 years, there were 1,850 CHD events, 1,075 stroke events, 2,908 HF cases, and 571 PAD cases. Gen 5 TnT showed a robust dose response association with each CVD type, with adjusted hazard ratio 2-4 for TnT ≥19 (vs.Table ). Even the category 6- Conclusions: Gen 5 TnT was highly correlated with Gen 4 TnT and gave quantifiable values in half middle-aged adults. Gen 5 TnT was robustly associated with major CVD events (especially HF and PAD) in the general population, supporting its usefulness in epidemiological research and clinical practice.

Introduction: Diabetes and metabolic syndrome (MetS) confer an increased risk of heart failure (HF) through poorly understood mechanisms. Galectin-3 (Gal-3) is a marker of fibrosis linked to greater HF risk. The inter-relationships among diabetes, MetS and Gal-3, and related implications for HF risk, are not well understood. Hypothesis: Diabetes and MetS are associated with elevated Gal-3, and high Gal-3 indicates greater HF risk among those with diabetes or MetS. Methods: We included 8,445 participants (mean age 63, 59% male, 21% Black) at ARIC Visit 4 (1996-1999) without baseline HF. We categorized participants by metabolic risk (no diabetes/no MetS; MetS only; diabetes with or without MetS), and Gal-3 levels (gender-specific quartiles). We assessed the associations of metabolic risk categories with high Gal-3 (≥75 th percentile) using logistic regression. We used Cox regression to evaluate associations of cross-categories of metabolic risk group and Gal-3 quartiles with incident HF. Results: In cross-sectional analyses, those with MetS were more likely to have elevated Gal-3 levels than those with no diabetes or MetS (OR 1.29, 95%CI 1.13-1.47). The additional presence of diabetes did not change the likelihood of elevated Gal-3 (OR 1.29, 95%CI 1.08-1.55). Over 20.5 years of follow-up, there were 1,611 HF events. Higher Gal-3 was associated with higher HF risk in each metabolic risk group, and higher metabolic risk group was associated with greater HF risk in each Gal-3 quartile. There was no interaction between Gal-3 and metabolic risk group on HF risk ( P =0.15). The combination of diabetes &MetS and high Gal-3 was associated with an almost 5-fold higher risk of incident HF (HR 4.93; 95% CI: 3.77 - 6.44) than the combination of no diabetes/MetS and low Gal-3 (first quartile) ( Table ). Conclusions: MetS was associated with higher levels of Gal-3. Metabolic risk group and Gal-3 provided powerful complementary prognostic information regarding HF risk. Fibrosis likely plays a role in the development of HF linked to metabolic risk.

Introduction: Growth differentiation factor 15 (GDF-15) is a marker of oxidative stress and inflammation and has been associated with several cardiovascular disease (CVD) phenotypes. However, conflicting results have been reported regarding the association of GDF-15 with incident atrial fibrillation (AF) in the general population. Hypotheses: Higher GDF-15 level is associated with increased risk of incident AF independent of potential confounders. Methods: In 10,101 White and Black ARIC participants (mean age 60 years and 20.9% Blacks) free of AF at baseline (1993-95), we quantified the association of GDF-15 and incident AF using three Cox proportional hazards models. GDF-15 was measured by SOMA scan assay. AF was defined by hospitalizations with AF diagnosis or death certificates (ICD-9 codes: 427.31-427.32; ICD-10 codes: I48.x) or AF diagnosis by ECG at subsequent ARIC visits. Results: There were 2165 cases of incident AF over a median follow-up of 20.7 years (incidence rate 12.1 cases/1,000 person-years). After adjusting for demographic characteristics and cardiovascular risk factors, log GDF-15 was significantly associated with incident AF (hazard ratio 1.42 (1.25-1.63) for top vs. bottom quartile) (Model 1 in Table ). The result was robust even further adjusting for history of other CVD phenotypes and cardiac markers (Models 2 and 3 in Table ). In Model 3, quartiles of high-sensitive cardiac troponin T (hs-cTnT) did not demonstrate significant associations with incident AF. Conclusions: In community-based population, elevated GDF-15 level was independently and robustly associated with incident AF (even more strongly than troponin). These results suggest the involvement of GDF-15 in the development of AF and the potential of GDF-15 as a risk marker to identify individuals at high risk of AF.

Background: Varicose veins are common in older adults, and associated with major clinical adverse outcomes such as deep venous thrombosis. Although there are a few established risk factors of varicose veins such as female sex, height, and obesity, some other risk factors demonstrated inconsistent results. Aim: To identify demographic and clinical risk factors of incident varicose veins Methods: Using data from the Atherosclerosis Risk in Communities (ARIC) Study over five clinic visits, we constructed a cohort of 6612 adults aged 65-70 years to be able to capture outpatient visits using Medicare data. For each participant, the first visit when meeting age range (65-70 years) was used as baseline. Varicose veins were defined as two outpatient encounters (at least a week apart) or inpatient diagnoses through 2015 (ICD9 codes:454.xx). Participants with clinically-recognized varicose veins prior to baseline were excluded. We explored demographic and clinical predictors routinely evaluated in ARIC visits using Cox regression. Results: During a median follow-up of 15 years, 348 (5%) of participants developed incident clinically-recognized varicose veins. We confirmed female sex, greater height, and higher body mass index were associated with varicose veins ( Table ). In addition, white race, loop diuretic use, prevalent heart failure were independently associated with incident varicosities. In subsample with relevant data, higher cardiac troponin T and natriuretic peptide were also independently associated with varicose veins. When we censored participants who left Medicare fee-for-service, results were largely consistent. Conclusions: In this community-based cohort study of older adults, in addition to established predictors, we found that white race and clinical and treatment factors related to cardiac function were associated with clinically-recognized varicose veins that may help identify those at high-risk of varicose veins.

Background: Greater central artery stiffness predicts cardiovascular disease and all-cause mortality, thus understanding arterial stiffness determinants has prevention implications. Reports of the temporal association of blood pressure with arterial stiffness are conflicting and the association with myocardial oxygen demand has not been evaluated. Objective: Characterize the association of mid- to later-life cumulative exposure to systolic blood pressure (SBP), myocardial oxygen demand, and hypertension (HTN) with arterial stiffness and its 5-year change in older adults. Methods: We included 1,975 adults (1151 women; 359 Black; visit 5 mean age 74 years) examined in visits 5 (2011-13) and 6 or 7 (2016-19) of the population-based ARIC-NCS with measures of arterial stiffness (carotid-femoral pulse wave velocity (cfPWV)). Higher cfPWV indicates greater arterial stiffness. We calculated cumulative exposures as the sum of averages from four consecutive visits from 1987-89 to 1996-98 divided by total time. Myocardial oxygen demand was calculated as the rate pressure product (RPP): (SBP x heart rate)/1,000. We derived HTN duration as the time since first HTN detection. Associations of cumulative exposures with visit 5 cfPWV and the 5-year cfPWV change were evaluated by multivariable linear regression adjusted for demographics and cardiometabolic factors. Results: Over the mean 5.7 years between visits 5 and 6 or 7, cfPWV increased by 144.9 cm/s (SD: 276.0; range -680.0, 961.5 cm/s). HTN at any visit, duration, and the time-weighted cumulative measures were associated with higher visit 5 cfPWV (Table). Prevalent HTN was inversely associated with cfPWV change. No statistically significant associations were observed for the other exposures and cfPWV change. Conclusion: Cumulative exposure to SBP, RPP, and HTN are modifiable traits associated with higher cfPWV at later-life, but not with rate of cfPWV change in older adulthood. HTN at visit 5 was associated with lower cfPWV change, albeit the change is of small magnitude.

Background: Sex hormones are associated with obesity, diabetes mellitus, and other stroke risk factors; however, studies on sex hormones and stroke risk report inconsistent results. We assessed the associations of testosterone and sex hormone binding globulin (SHBG) with risk of ischemic stroke among men and post-menopausal women in the Atherosclerosis Risk in Communities (ARIC) Study. Methods: A total of 4,349 men and 4,720 post-menopausal women who had SHBG and total testosterone measurements at visit 4 (1996-98) were followed through 2018 for the development of ischemic stroke. We examined log transformed SHBG and testosterone exposure as quartiles and as per 1 SD increment. We used Cox regression to estimate the hazard ratios (HR) for ischemic stroke, adjusting for demographic, behavioral and clinical variables. Analyses were stratified by sex and menopausal hormone therapy (HT) use. Results: Participants were aged 63±6 years at baseline, 52% were women (25% HT users), and 21% Black. There were 691 strokes over a median follow-up of 19.8 years. Mean log SHBG (nmol/L) and testosterone (nmol/L) were 4.3±0.7 and 3.1±0.5 for HT users, 3.6±0.7 and 3.2±0.5 for non-HT users, and 3.4±0.5 and 6.2±0.5 for men. Quartile 1 vs Q4 for SHBG and testosterone were ≤50.3 vs >121 and ≤16 vs >28 in HT users, ≤23.3 vs >55 and ≤17.9 vs >32.7 in non-HT users, and ≤23.3 vs >41.8 and ≤388 vs >657 in men. SHBG and testosterone were not significantly associated with stroke in any group (Figure). The HRs [95% confidence interval] for highest to lowest SHBG and testosterone quartiles were 1.04 [0.51-2.14] and 1.64 [0.85-3.17] in HT-users, 1.02 [0.70-1.48] and 1.16 [0.83-1.62] in HT non-users, and 0.96 [0.70-1.32] and 0.87 [0.63-1.21] in men, respectively. The HRs for stroke associated with 1 SD increase in SHBG and testosterone were 1.14 [0.88-1.46] and 1.19 [0.96-1.46] in HT users. Associations were also null among non-HT users and men. Conclusion: SHBG and testosterone were not associated with ischemic stroke risk in this cohort of older men and post-menopausal women.

Background: In the 2018 AHA/ACC Cholesterol guideline, risk stratification is an essential element. The use of a Pooled Cohort Equation (PCE) is recommended for individuals without atherosclerotic cardiovascular disease (ASCVD), and the new dichotomous classification of very high-risk vs. high-risk has been introduced for patients with ASCVD. These distinct risk stratification systems mainly rely on traditional risk factors, raising the possibility that a single model can predict major adverse cardiovascular events (MACEs) in persons with and without ASCVD. Methods: We studied 11,335 ARIC participants with (n=885) and without (n=10,450) a history of ASCVD (myocardial infarction, ischemic stroke, and symptomatic peripheral artery disease) at baseline (1996-98). We modeled factors in the PCE and the new classification for ASCVD patients (Figure legend) in a single CVD prediction model. We examined their associations with MACEs (myocardial infarction, stroke, and heart failure) using Cox models and evaluated the discrimination and calibration for a single model including those factors. Results: During a median follow-up of 18.4 years, there were 3,658 MACEs (3,105 in participants without ASCVD). In general, the factors in the PCE and the risk classification system for ASCVD patients were associated similarly with MACEs regardless of baseline ASCVD status, although age and systolic blood pressure showed significant interactions. A single model with these predictors and the relevant interaction terms showed good calibration and discrimination for those with and without ASCVD (c-statistic=0.729 and 0.704, respectively) (Figure). Conclusion: A single CVD prediction model performed well in persons with and without ASCVD. This approach will provide a specific predicted risk to ASCVD patients (instead of dichotomy of very high vs. high risk) and eliminate a practice gap between primary vs. secondary prevention due to different risk prediction tools.

Objective: The aim of this study was to comprehensively assess the association of multiple lipid measures with incident peripheral artery disease (PAD). Approach and Results: We used Cox proportional hazards models to characterize the associations of each of the fasting lipid measures (total cholesterol, LDL-C [low-density lipoprotein cholesterol], HDL-C [high-density lipoprotein cholesterol], triglycerides, RLP-C [remnant lipoprotein cholesterol], LDL-TG [LDL-triglycerides], sdLDL-C [small dense LDL-C], and Apo-E-HDL [Apo-E-containing HDL-C]) with incident PAD identified by pertinent International Classification of Diseases, Ninth Revision, Clinical Modification ( ICD-9-CM ) hospital discharge codes (eg, 440.2) among 8330 Black and White ARIC (Atherosclerosis Risk in Communities) participants (mean age 62.8 [SD 5.6] years) free of PAD at baseline (1996–1998) through 2015. Since lipid traits are biologically correlated to each other, we also conducted principal component analysis to identify underlying components for PAD risk. There were 246 incident PAD cases with a median follow-up of 17 years. After accounting for potential confounders, the following lipid measures were significantly associated with PAD (hazard ratio per 1-SD increment [decrement for HDL-C and Apo-E-HDL]): triglycerides, 1.21 (95% CI, 1.08–1.36); RLP-C, 1.18 (1.08–1.29); LDL-TG, 1.18 (1.05–1.33); HDL-C, 1.39 (1.16–1.67); and Apo-E-HDL, 1.27 (1.07–1.51). The principal component analysis identified 3 components (1: mainly loaded by triglycerides, RLP-C, LDL-TG, and sdLDL-C; 2: by HDL-C and Apo-E-HDL; and 3: by LDL-C and RLP-C). Components 1 and 2 showed independent associations with incident PAD. Conclusions: Triglyceride-related and HDL-related lipids were independently associated with incident PAD, which has implications on preventive strategies for PAD. However, none of the novel lipid measures outperformed conventional ones. Graphic Abstract: A graphic abstract is available for this article.

Aims Despite statin and antihypertensive therapies, older Americans have high atherosclerotic cardiovascular disease (ASCVD) risk. Novel measures of triglyceride-rich lipoproteins, low-density lipoprotein triglycerides (LDL-TG), and remnant-like particle cholesterol (RLP-C), are associated with ASCVD in middle-aged adults. Polymorphisms in genes encoding angiopoietin-related protein 3 (ANGPTL3) and apolipoprotein C-III (apoC-III), two proteins involved in triglyceride catabolism, are associated with increased risk for hypertriglyceridaemia and ASCVD and are potential therapeutic targets. We examined associations of LDL-TG, RLP-C, apoC-III, and ANGPTL3 levels with ASCVD events in older adults in the Atherosclerosis Risk in Communities (ARIC) study. Methods and results In 6359 participants (mean age 75.8 ± 5.3 years) followed for ASCVD events [coronary heart disease (CHD) or ischaemic stroke] up to 6 years, associations between LDL-TG, RLP-C, apoC-III, and ANGPTL3 and ASCVD events were assessed using Cox regression. With adjustment for age, sex, and race, RLP-C, LDL-TG, apoC-III, and ANGPTL3 (as continuous variables) were significantly associated with CHD. However, after adjustment for traditional risk factors and lipid-lowering medications, only LDL-TG and ANGPTL3 were significantly associated with ASCVD events [hazard ratio (HR) 1.72, 95% confidence interval (CI) 1.25–2.37 per log unit increase in LDL-TG; HR 1.63, 95% CI 1.17–2.28 per log unit increase in ANGPTL3]. Conclusions In older adults, LDL-TG, RLP-C, apoC-III, and ANGPTL3 were associated with CHD events in minimally adjusted models; LDL-TG and ANGPTL3 remained independent predictors of ASCVD events with further adjustment. Future studies should assess potential benefit of lowering hepatic apoC-III or ANGPTL3 expression in patients with elevated triglyceride-rich lipoproteins.

Importance: Human genetics and studies in experimental models support a key role of monocyte-chemoattractant protein-1 (MCP-1) in atherosclerosis. Yet, the associations of circulating MCP-1 levels with risk of coronary heart disease and cardiovascular death in the general population remain largely unexplored. Objective: To explore whether circulating levels of MCP-1 are associated with risk of incident coronary heart disease, myocardial infarction, and cardiovascular mortality in the general population. Data Sources and Selection: Population-based cohort studies, identified through a systematic review, that have examined associations of circulating MCP-1 levels with cardiovascular end points. Data Extraction and Synthesis: Using a prespecified harmonized analysis plan, study-specific summary data were obtained from Cox regression models after excluding individuals with overt cardiovascular disease at baseline. Derived hazard ratios (HRs) were synthesized using random-effects meta-analyses. Main Outcomes and Measures: Incident coronary heart disease (myocardial infarction, coronary revascularization, and unstable angina), nonfatal myocardial infarction, and cardiovascular death (from cardiac or cerebrovascular causes). Results: The meta-analysis included 7 cohort studies involving 21401 individuals (mean [SD] age, 53.7 [10.2] years; 10012 men [46.8%]). Mean (SD) follow-up was 15.3 (4.5) years (326392 person-years at risk). In models adjusting for age, sex, and race/ethnicity, higher MCP-1 levels at baseline were associated with increased risk of coronary heart disease (HR per 1-SD increment in MCP-1 levels: 1.06 [95% CI, 1.01-1.11]; P =.01), nonfatal myocardial infarction (HR, 1.07 [95% CI, 1.01-1.13]; P =.02), and cardiovascular death (HR, 1.12 [95% CI, 1.05-1.20]; P

Membrane-bound angiotensin-converting enzyme 2 is important in regulation of the renin–angiotensin–aldosterone system, but the association of cleaved soluble ACE2 (sACE2) with cardiovascular disease (CVD) is unclear. We evaluated the association of sACE2 with cardiac biomarkers, structure, and function and cardiovascular events in the Atherosclerosis Risk in Communities Study. sACE2 was measured in a subset of 497 participants (mean age 78±5.4 years, 53% men, 27% black); Cox regression analyses assessed prospective associations of sACE2 with time to first CVD event at median 6.1-year follow-up. sACE2 was higher in men, blacks, and participants with prevalent CVD, diabetes, or hypertension. Higher sACE2 levels were associated with significantly higher biomarkers of cardiac injury (high-sensitivity cardiac troponin I and T, N-terminal pro–B-type natriuretic peptide), greater left ventricular mass index, and impaired diastolic function in linear regression analyses, and with increased risk for heart failure hospitalization (adjusted hazard ratio per natural log unit increase [HR] 1.32, 95% confidence interval [CI] 1.10 to 1.58), CVD events (HR 1.34, 95% CI 1.13 to 1.60), and all-cause death (HR 1.26, 95% CI 1.01 to 1.57). In an elderly biracial cohort, sACE2 was positively associated with biomarkers reflecting myocardial injury and neurohormonal activation, left ventricular mass index, impaired diastolic function, CVD, events and all-cause death.

Background: Few studies have evaluated the link between trace or mild valvular disease with measures of cardiac structure, function, and damage in a general population of older adults. Hypothesis: Three left-sided valvular conditions (aortic stenosis, aortic regurgitation, and mitral regurgitation) will be independently associated with cardiac remodeling, damage, and overload. Methods: In 4,935 ARIC participants aged 66-99 years in 2011-13, we examined the cross-sectional associations of these three valvular conditions, only in trace or mild forms, with echocardiographic measures (left ventricular mass index, left ventricular end-diastolic diameter [LVDd], ejection fraction [EF], left atrial volume index [LAVI]) and biomarkers (high sensitivity troponin-T [hs-TnT] and natriuretic peptide) using multivariable linear regression. Aortic stenosis was categorized as none or mild by peak transaortic jet velocity and mean transaortic gradient. Regurgitation was categorized as none, trace, or mild based on color Doppler signal (see the Table footnote for detailed definitions). Results: The prevalence was 4.3% for mild aortic stenosis, 10.8% for aortic regurgitation (10.3% trace, 0.5% mild), and 44.0% for mitral regurgitation (39.9% trace, 4.1% mild). Each valvular condition showed independent and graded associations with all measures tested (Table), with the exception of aortic stenosis with LVDd and aortic regurgitation with hs-TnT. There was a positive association between aortic stenosis and EF. The associations remained consistent when all three valvular conditions were modeled simultaneously. Conclusions: Three prevalent valvular conditions were independently associated with cardiac remodeling, damage, and overload. Although this study cannot determine the directionality of these associations, our results suggest the involvement of mild valvular abnormalities in the pathophysiology of functional and structural alteration of the heart.

Introduction: Membrane-bound angiotensin-converting enzyme 2 (ACE2) has been identified to a have prominent role in SARS-COV-2 infection and is an important counter-regulator of renin-angiotensin system. But the association of cleaved soluble ACE2 (sACE2) with cardiovascular disease (CVD) remains unclear. We sought to identify the association of sACE2 with cardiac biomarkers, structure, function and events in the Atherosclerosis Risk in Communities (ARIC) Study. Methods: sACE2 was measured in a subset of 497 patients from a case-control study of incident heart failure (HF) at visit 5 (2011-2013), mean age 78 (SD 5.4), 53% men and 27% black. We used linear regression to evaluate the associations of sACE2 with cardiac biomarkers (hs-cTnI, hs-cTnT, NT-proBNP) and echocardiographic parameters. We used Cox regression to evaluate associations of sACE2 with risk of HF hospitalization, global CVD events (CHD, ischemic stroke or HF hospitalization) and all-cause death. Results: Over a median follow up of 6.1 (4.6, 6.8) years, 282 global CVD events and 190 all-cause deaths occurred. sACE2 levels were higher in men, blacks, those with prevalent CVD, diabetes and hypertension. Higher sACE2 levels were associated with significantly higher hs-cTnI, hs-cTnT, NT-proBNP levels, greater left ventricular (LV) mass index, impaired diastolic function ( Table ) and increased risk for HF hospitalization (adjusted HR 1.32 per log unit increase, 95% CI 1.10-1.58), global CVD events (HR 1.34, 95% CI 1.13-1.60) and all-cause death (HR 1.26, 95% CI 1.01-1.57). Conclusions: In an elderly biracial cohort, sACE2 was positively associated with biomarkers reflecting myocardial injury and neurohormonal activation, LV mass index, impaired diastolic function, CVD events and all-cause death. Future research is needed to elucidate the significance of sACE2 in development of CVD, not only in patients with SAR—COV2 infection, but the general population

Introduction: Clinical trials suggest that low-saturated fat diet (LSFD) may reduce the risk of atherosclerotic cardiovascular disease (ASCVD) in subjects with hypertriglyceridemia (HTG). Monocytes play crucial roles in atherogenesis. Hypothesis: LSFD vs high-saturated fat diet (HSFD) improves monocyte phenotypes, thereby reducing ASCVD risk, in subjects with HTG. Methods: Subjects with HTG and metabolic syndrome (MetS, n=19) received isocaloric LSFD (~25% of calories from fat, 5% from saturated fat) and HSFD (~52% of calories from fat, 25% from saturated fat) in randomized order for 4 days (days 1-4) plus a breakfast on day 5, separated by a 4- to 6-week washout period. Blood was drawn on day 1 fasting before the diets and 3 times on day 5 (fasting before the breakfast and 4 and 6 hours postprandial) for measurement of lipid profile and analyses of monocyte phenotypes by flow cytometry and monocyte adhesion by a lab-on-a-chip microfluidic assay. Results: On day 5, LSFD, compared to HSFD, induced lower plasma levels of postprandial total triglyceride and LDL-triglyceride and fasting and postprandial total cholesterol, LDL-cholesterol, and small dense LDL-cholesterol. Compared to HSFD, LSFD reduced fasting and postprandial intracellular lipid accumulation in classical and intermediate monocytes examined by nile red staining and indicated by side scatter value of flow cytometric analysis. LSFD versus HSFD also reduced ex vivo uptake of oxidized LDL by classical monocytes at 4 hours postprandially and by intermediate monocytes in fasting state. Surface levels of molecules involved in monocyte adhesion/migration, including CD11c, CD81, and CCR2, were lower on monocytes with LSFD than with HSFD. Consistently, LSFD compared to HSFD reduced monocyte adhesion to VCAM-1. Intracellular levels of cytokines such as IL-1β, TNFα, and IL-6 in monocytes showed no difference between the two diets. Conclusions: In subjects with HTG and MetS, short-term LSFD compared to HSFD reduces monocyte intracellular lipid accumulation and improves monocyte phenotypes with reductions in monocyte adhesion and oxidized LDL uptake. These findings highlight the importance of diet composition in monocyte phenotypes and possibly atherosclerosis risks in patients with HTG and MetS.

I ntroduction: Systolic blood pressure (SBP) is an important component of all cardiovascular disease (CVD) risk prediction equations but its biological variability and impact on estimated risk is a concern. Furthermore, predictive value of SBP may differ in older individuals where traditional risk factors (TRF) are less predictive. Hypothesis: Biomarkers reflecting hypertension-related end organ injury (hsTnT, NT-proBNP, eGFR), improve CVD risk prediction in older but not middle age adults as compared to SBP. Methods: Using data from visits 2 (1990-92) and 5 (2011-13) of ARIC, we developed 3 models- Model 1 included all TRF; Model 2- all TRF except SBP + individual biomarkers and Model 3 all TRF + individual biomarkers. C-statistics were used to assess risk discrimination for coronary heart disease, stroke, heart failure, and CVD. Results: After excluding those with prevalent CVD, there were 12,567 individuals at visit 2 (mean age 57, SD 6 years; 43% men) and 4,508 individuals at visit 5 (mean age 76, SD 5 years; 37% men). Over a median (IQR) follow-up time of 22 (12.4–26.7) years and 6.2 (5.4–6.8) years, the incidence rates of CVD events (per 1000 person-years) were 19.0 and 21.8 at visits 2 and 5, respectively. At visit 2, the model with SBP and biomarkers resulted in the largest improvement in C-statistic and SBP contributed to all models. However, at visit 5, removing SBP from the models with the biomarkers had no impact on C-statistic while the addition of the biomarkers (especially hsTnT and NT-proBNP) significantly improved C-statistics for most outcomes ( Table ). Among the biomarkers eGFR had the least additive value. Conclusions: HsTnT and NT-proBNP significantly improve risk discrimination of CHD, stroke, and HF among middle and older adults, while SBP has value in middle age but not in older age. Biomarkers should be considered in risk prediction equations in older individuals where the value of TRF such as SBP decrease.

Introduction: Cardiovascular disease (CVD) is the most common cause of death in nonalcoholic steatohepatitis (NASH). While these conditions share many cardio-metabolic risk factors including metabolic syndrome, diabetes and dyslipidemia, limited data exist on whether NASH is independently and prospectively associated with incident CVD beyond traditional risk factors. Fibrosis-4 (FIB-4) index is a scoring system based on platelet count, age, AST and ALT, shown to be comparable to magnetic resolution elastography for predicting advanced fibrosis in biopsy-proven NASH. We sought to evaluate the association of elevated FIB-4 with global CVD events and CVD mortality in the Atherosclerosis Risk in Communities (ARIC) Study Methods: We studied 5531 individuals, mean age of 76 (SD 5.2) years, 58% female, 22% black, at ARIC visit 5 (2011-2013). FIB-4 was categorized as low risk of advanced fibrosis for score 3.25. Cox regression was used to estimate the association of FIB-4 with time to first global CVD event (CHD, ischemic stroke or heart failure hospitalization) and CVD mortality adjusted for pooled cohort equation risk factors. Results: Over a median follow up of 6.2 (5.3-6.8) years, there were 1108 global CVD events and 457 CVD deaths. In adjusted models, compared to participants with low FIB-4 (3.25, had significantly increased risk for global CVD events (HR 1.58, 95% CI 1.23-2.02) and CVD mortality (HR 1.70, 95% CI 1.16-2.50). Conclusions: In a large prospective cohort, presence of advanced liver fibrosis, as assessed by elevated FIB-4 index >3.25, was associated with increased risk for CVD events and CVD mortality, beyond traditional CVD risk factors. Future clinical trials of candidate medications under study for NASH should examine whether effective NASH treatment will impact CV outcomes.

Objective: Inflammatory markers, such as hs-CRP (high-sensitivity C-reactive protein), have been reported to be related to peripheral artery disease (PAD). Galectin-3, a biomarker of fibrosis, has been linked to vascular remodeling and atherogenesis. However, its prospective association with incident PAD is unknown; as is the influence of inflammation on the association between galectin-3 and PAD. Approach and Results: In 9851 Atherosclerosis Risk in Communities Study participants free of PAD at baseline (1996–1998), we quantified the association of galactin-3 and hs-CRP with incident PAD (hospitalizations with PAD diagnosis [ International Classification of Diseases - Ninth Revision : 440.2–440.4] or leg revascularization [eg, International Classification of Diseases - Ninth Revision : 38.18]) as well as its severe form, critical limb ischemia (PAD cases with resting pain, ulcer, gangrene, or leg amputation) using Cox models. Over a median follow-up of 17.4 years, there were 316 cases of PAD including 119 critical limb ischemia cases. Log-transformed galectin-3 was associated with incident PAD (adjusted hazard ratio, 1.17 [1.05–1.31] per 1 SD increment) and critical limb ischemia (1.25 [1.05–1.49] per 1 SD increment). The association was slightly attenuated after further adjusting for hs-CRP (1.14 [1.02–1.27] and 1.22 [1.02–1.45], respectively). Log-transformed hs-CRP demonstrated robust associations with PAD and critical limb ischemia even after adjusting for galectin-3 (adjusted hazard ratio per 1 SD increment 1.34 [1.18–1.52] and 1.34 [1.09–1.65], respectively). The addition of galectin-3 and hs-CRP to traditional atherosclerotic predictors (C statistic of the base model 0.843 [0.815–0.871]) improved the risk prediction of PAD (ΔC statistics, 0.011 [0.002–0.020]). Conclusions: Galectin-3 and hs-CRP were independently associated with incident PAD in the general population, supporting the involvement of fibrosis and inflammation in the pathophysiology of PAD.

Background Circulating galectin‐3 levels provide prognostic information in patients with established heart failure (HF), but the associations between galectin‐3 levels and other incident cardiovascular events in asymptomatic individuals at midlife and when remeasured ≈15 years later are largely uncharacterized. Methods and Results Using multivariable Cox proportional hazards models, we identified associations between plasma galectin‐3 levels (hazard ratio [HR] per 1 SD increase in natural log galectin‐3) and incident coronary heart disease, ischemic stroke, HF hospitalization, and total mortality in ARIC (Atherosclerosis Risk in Communities) participants free of cardiovascular disease at ARIC visit 4 (1996–1998; n=9247) and at ARIC visit 5 (2011–2013; n=4829). Higher galectin‐3 level at visit 4 (median age 62) was independently associated with incident coronary heart disease (adjusted HR, 1.30; 95% CI, 1.06–1.60), ischemic stroke (HR, 1.42; 95% CI, 1.01–2.00), HF (HR, 1.44; 95% CI, 1.17–1.76), and mortality (HR, 1.56; 95% CI, 1.35–1.80). At visit 5 (median age, 74), higher galectin‐3 level was associated with incident HF (HR, 1.93; 95% CI, 1.15–3.24) and total mortality (HR, 1.70; 95% CI, 1.15–2.52), but not coronary heart disease or stoke. Individuals with the greatest increase in galectin‐3 levels from visit 4 to visit 5 were also at increased risk of incident HF and total mortality. Conclusions In a large, biracial community‐based cohort, galectin‐3 measured at midlife and older age was associated with increased risk of cardiovascular events. An increase in galectin‐3 levels over this period was also associated with increased risk.

Background Elevated lipoprotein(a) (Lp[a]) and family history (FHx) of coronary heart disease (CHD) are individually associated with cardiovascular risk, and Lp(a) is commonly measured in those with FHx. Objectives The aim of this study was to determine independent and joint associations of Lp(a) and FHx with atherosclerotic cardiovascular disease (ASCVD) and CHD among asymptomatic subjects. Methods Plasma Lp(a) was measured and FHx was ascertained in 2 cohorts. Elevated Lp(a) was defined as the highest race-specific quintile. Independent and joint associations of Lp(a) and FHx with cardiovascular risk were determined using Cox regression models adjusted for cardiovascular risk factors. Results Among 12,149 ARIC (Atherosclerosis Risk In Communities) participants (54 years, 56% women, 23% black, 44% with FHx), 3,114 ASCVD events were observed during 21 years of follow-up. FHx and elevated Lp(a) were independently associated with ASCVD (hazard ratio [HR]: 1.17; 95% confidence interval [CI]: 1.09 to 1.26, and HR: 1.25; 95% CI: 1.12 to 1.40, respectively), and no Lp(a)-by-FHx interaction was noted (p = 0.75). Compared with subjects without FHx and nonelevated Lp(a), those with either elevated Lp(a) or FHx were at a higher ASCVD risk, while those with both had the highest risk (HR: 1.43; 95% CI: 1.27 to 1.62). Similar findings were observed for CHD risk in ARIC, in analyses stratified by premature FHx, and in an independent cohort, the DHS (Dallas Heart Study). Presence of both elevated Lp(a) and FHx resulted in greater improvement in ASCVD and CHD risk reclassification and discrimination indexes than either marker alone. Conclusions Elevated plasma Lp(a) and FHx have independent and additive joint associations with cardiovascular risk and may be useful concurrently for guiding primary prevention therapy decisions.

Rationale: Proinflammatory cytokines have been identified as potential targets for lowering vascular risk. Experimental evidence and Mendelian randomization suggest a role of MCP-1 (monocyte chemoattractant protein-1) in atherosclerosis and stroke. However, data from large-scale observational studies are lacking. Objective: To determine whether circulating levels of MCP-1 are associated with risk of incident stroke in the general population. Methods and Results: We used previously unpublished data on 17 180 stroke-free individuals (mean age, 56.7±8.1 years; 48.8% men) from 6 population-based prospective cohort studies and explored associations between baseline circulating MCP-1 levels and risk of any stroke, ischemic stroke, and hemorrhagic stroke during a mean follow-up interval of 16.3 years (280 522 person-years at risk; 1435 incident stroke events). We applied Cox proportional-hazards models and pooled hazard ratios (HRs) using random-effects meta-analyses. After adjustments for age, sex, race, and vascular risk factors, higher MCP-1 levels were associated with increased risk of any stroke (HR per 1-SD increment in ln-transformed MCP-1, 1.07; 95% CI, 1.01–1.14). Focusing on stroke subtypes, we found a significant association between baseline MCP-1 levels and higher risk of ischemic stroke (HR, 1.11 [1.02–1.21]) but not hemorrhagic stroke (HR, 1.02 [0.82–1.29]). The results followed a dose-response pattern with a higher risk of ischemic stroke among individuals in the upper quartiles of MCP-1 levels as compared with the first quartile (HRs, second quartile: 1.19 [1.00–1.42]; third quartile: 1.35 [1.14–1.59]; fourth quartile: 1.38 [1.07–1.77]). There was no indication for heterogeneity across studies, and in a subsample of 4 studies (12 516 individuals), the risk estimates were stable after additional adjustments for circulating levels of IL (interleukin)-6 and high-sensitivity CRP (C-reactive protein). Conclusions: Higher circulating levels of MCP-1 are associated with increased long-term risk of stroke. Our findings along with genetic and experimental evidence suggest that MCP-1 signaling might represent a therapeutic target to lower stroke risk.Visual Overview: An online visual overview is available for this article.

Background: Minute elevations in cardiac troponins detected with high sensitivity assays are associated with cardiovascular disease (CVD) and mortality. The biology underlying these associations is uncertain. We identified plasma proteins associated with cardiac troponins and assessed the relation of troponin-associated proteins with CVD and all-cause mortality in adults without prevalent CVD. Methods: We included 3451 adults (62% female, 18% black, mean age: 75 years) in ARIC at Visit 5 (2011-2013). High sensitivity assays were used to measure troponins (hs-cTnI [Abbott], hs-cTnT [Roche]). We used linear regression to model the association of log2-transformed troponin and plasma proteins (SomaLogic aptamer assay). Upstream regulators of the troponin-associated proteins were identified using Ingenuity Pathway Analysis. Elastic net was used to extract troponin-associated protein signatures for principal component analysis and Cox models of CVD and all-cause mortality (median follow-up: 6 years). Results: Of the 4326 plasma proteins, there were 281 significantly (p Conclusions: Hs-cTnI and hs-cTnT have overlapping, but distinct, proteomic signatures. The underlying proteomic signature explains most of the mortality risk associated with hs-cTnT, but not hs-cTnI.

Background: Cerebrovascular disease is often the consequence of cardiac disease. Our aim was to examine associations of biomarkers of cardiovascular disease, high sensitivity troponin T (hs-cTnT), NT-proBNP, and galectin-3, with cerebrovascular signs: lacunar infarcts, lobar and subcortical microhemorrhages, cortical infarcts, and white matter hyperintensity (WMH) volume. We also examined total cortical and Alzheimer’s Disease (AD) signature region volumes. Methods: We conducted a cross-sectional analysis of 1748 ARIC participants from the 2011-2013 exam who had biomarker measurements, completed a brain MRI, and did not have a clinical history of stroke. We used linear regression to model brain volumes, modeled as Z scores, and logistic regression for all other outcomes; biomarkers were log transformed. We repeated analyses excluding persons with coronary heart disease, atrial fibrillation, and heart failure. Results: The mean age of participants was 76, 62% were female, and 21% were Black. All biomarkers were associated with total cortical volume. Each standard deviation increase in log hs-cTnT was associated with lower total cortical volume (adjusted beta = -0.08, 95% CI: -0.12, -0.05); results for the other biomarkers were similar (Figure). All biomarkers were associated with lobar microhemorrhages. Hs-cTnT and NT-proBNP were associated with WMH volume, but galectin-3 was not. No biomarker was associated with subcortical microhemorrhages or cortical infarcts. Results were similar in persons without coronary heart disease, atrial fibrillation, or heart failure (conditions associated with cerebral thromboembolism). Conclusions: In persons free of clinical cardiovascular disease, biomarkers of cardiac stretch, strain, and fibrosis were associated cerebral small vessel disease and reduced cortical volume, but not in a specific pattern suggestive of AD pathogenesis. This suggests subclinical vascular insults affect brain structure through mixed pathogenic processes.

Introduction: Mechanistic studies suggest an involvement of growth differentiation factor 15 (GDF-15) in metabolic dysregulation. However, the potential utility of GDF-15 as a marker of diabetes or metabolic syndrome (MetS) risk remains unclear, especially in older adults. Hypothesis: GDF-15 is positively associated with biomarkers of hyperglycemia, diabetes, and MetS. Methods: We conducted a cross-sectional analysis of older adults who attended visit 6 (2016-2017) of the Atherosclerosis Risk in Communities (ARIC) Study. GDF-15 was measured using electrochemiluminescence immunoassay (Elecsys, Roche Diagnostics). Linear regression was used to assess continuous outcomes after appropriate transformations, and multivariable-adjusted odds of diabetes or MetS by quartiles of GDF-15 were derived using logistic regression. Results: Among 3,792 participants (mean age 80 years, 59% women, 23% blacks and 77% whites), higher GDF-15 concentrations (per 1-unit increase in ln[GDF-15]) were associated with higher levels of fasting plasma glucose (mg/dL) (adjusted β coefficient : 10.98, 95% CI:8.86 - 13.09) and HbA 1C (%)(0.41, 95% CI: 0.35 - 0.48). Higher GDF-15 was associated with greater odds of diabetes (adjusted odds ratio [OR]: 5.81 for highest vs. lowest GDF-15 quartile, 95% CI 4.43-7.61) and of MetS syndrome (adjusted OR: 2.57, 95% CI 2.01-3.20) among individuals without diabetes (Figure). Conclusions: In this sample of older adults, elevated GDF-15 was strongly associated with diabetes and metabolic syndrome. These data strongly suggest that GDF-15 could be a robust biomarker of adverse metabolic states.

Introduction: There is growing evidence that high sensitivity troponin I (hs-cTnI) and high sensitivity troponin T (hs-cTnT) are associated with distinct clinical characteristics and perform differently for cardiovascular risk stratification. The comparative associations of hs-cTnI and hs-cTnT with microvascular complications, such as peripheral neuropathy (PN), have not been quantified. Methods: We conducted a cross-sectional analysis of 2,974 black and white participants in the ARIC study who underwent monofilament PN testing at visit 6 (2016-2017, age 71-94 years). Hs-cTnI and hs-cTnT were measured in serum and categorized according to population tertiles. We used logistic regression to assess the associations of hs-cTnI and hs-cTnT with PN among adults with and without diabetes after adjusting for traditional risk factors. Results: Overall, 38% of participants had evidence of PN. Median hs-cTnI and hs-cTnT levels were 3.3ng/L (IQR 2.2-5.2) and 11.0ng/L (8.0-16.0) for participants with PN and 2.6ng/L (1.8-4.0) and 8.0ng/L (6.0-12.0) for participants without PN, respectively. Only hs-cTnT was significantly associated with PN after adjusting for traditional risk factors ( Figure, panel A ) and after mutually adjusting for both troponin measurements ( Figure, panel B ). The association of hs-cTnT with PN was similar for adults with and without diabetes (P=0.62 for interaction). Conclusions: PN was common among older adults with and without diabetes. Hs-cTnI and hs-cTnT had distinct associations with PN, with a more robust association for hs-cTnT. These findings add to the growing body of knowledge regarding differences in the use of high sensitivity troponin assays for monitoring risk in the general population.

Introduction: The association of conventional lipids (total cholesterol, low-density lipoprotein cholesterol [LDL-C], high-density lipoprotein cholesterol [HDL-C], and triglycerides [TG]) with incident peripheral artery diseases (PAD) has been reported. Understanding novel lipids (remnant lipoprotein cholesterol [RLP-C], LDL-TG, small dense LDL-C [sdLDL-C], Apo-E containing HDL [ApoE-HDL]) in this regard could inform risk classification and preventive approaches for PAD. Hypothesis: Some novel lipids will be more evidently related to incident PAD than some conventional lipid measures. Methods: In 8808 ARIC participants (mean age 62.6 [SD 5.7] y) free of PAD at baseline (visit 4, 1996-98), we assessed the associations of each lipid measure (as quartiles and continuous variable) and incident PAD-related hospitalization through 2015 using multivariable Cox models. Since some lipids are closer biologically correlated than others, indicating collective effects among lipids, we used factor analysis to identify underlying factors and modeled them for PAD risk. Results: There were 262 incident PAD cases. After accounting for potential confounders, HDL-C, TG, RLP-C, LDL-TG, and ApoE-HDL were independently associated with incident PAD, with the largest hazard ratio between Q4 and Q1 for HDL-C (Table). LDL-C showed a J-shaped association, and Q4 and Q3 became significant when compared to Q2 (1.57 [1.09-2.25] and 1.46 [1.02-2.11], respectively). The associations of LDL-TG, and ApoE-HDL remained significant after adjusting for LDL-C. The factor analysis identified 3 factors (Factor 1: mainly loaded by TG, RLC, LDL-TG and sdLDL-C; Factor 2: by total cholesterol, LDL-C, and sdLDL-C; Factor 3: by HDL-C and ApoE-HDL), and only Factor 3 showed an independent association with incident PAD (0.78 [0.65-0.93]). Conclusions: These findings suggest the importance of HDL-related and TG-related lipids in the development of PAD, with potential therapeutic implications.