Multi-Ethnic Study Research Articles

The epidemiology of acne vulgaris in a multiethnic adolescent population from Rotterdam, the Netherlands: A cross-sectional study

Although acne is a prevalent multifactorial inflammatory skin condition, few studies were performed in multiethnic populations. To study the prevalence and determinants of acne in a multiethnic study at the start of puberty. This cross-sectional study is embedded in Generation R, a population-based prospective study from Rotterdam, the Netherlands. Three-dimensional facial photos at the center visit in 2016-2019 (of ∼13-year-olds) were used to grade acne severity using the Global Evaluation of the Acne Severity (GEA). Analyses were stratified by biological sex and explored through chi-square tests and multivariable ordinal logistic regression. A total of 4561 children (51% girls) with a median age of 13.5 (IQR 13.3-13.6) were included. The visible acne prevalence (GEA 2-5) for girls vs boys was 62% vs 45% and moderate-to-severe acne (GEA 3-5) 14% vs 9%. Higher puberty stages (adjusted odds ratios: 1.38 [1.20-1.59] and 2.16 [1.86-2.51] for girls and boys, respectively) and darker skin colors V and VI (adjusted odds ratios: 1.90 [1.17-3.08] and 2.43 [1.67-3.56]) were associated with more severe acne in both sexes, and being overweight in boys (adjusted odds ratio: 1.58 [1.15-2.17]). Cross-sectional design. Acne prevalence was high at the age of 13years and was associated with advanced puberty, darker skin color, and weight status.

Nonlinear associations between computed tomography-measures of adiposity and long pentraxin-3 in the Multi-Ethnic Study of Atherosclerosis.

Long pentraxin-3 (PTX-3) is an acute phase protein associated with cardiovascular disease, lung injury, and mortality. We evaluated the association between computed tomography (CT)-measurements of adipose tissue and plasma levels of PTX-3. We performed a cross-sectional analysis of community-dwelling adults enrolled in the multi-center Multiethnic Study of Atherosclerosis who underwent cardiac or abdominal CT and had available PTX-3 measurements. There was a U-shaped association between pericardial adipose tissue volume (PAT), abdominal visceral adipose tissue area (VAT), hepatic attenuation, and PTX-3 levels, with extremes of adiposity associated with greater PTX-3 levels. Using multivariable-adjusted piecewise regression models, among participants with low PAT, every 1% increase in PAT volume was associated with a 13.8% decrease in PTX-3 (95% confidence interval [CI] -21.6 to -6.0); among participants with high PAT, every 1% increase in PAT volume was associated with a 6.0% increase in PTX-3 (95% CI -0.4 to 12.5). Results were similar for abdominal VAT and hepatic attenuation. In a cohort of community-dwelling adults, we demonstrated a "U-shaped" association between pericardial, abdominal visceral, and hepatic adiposity with PTX3 levels, suggesting that extreme adiposity is associated with greater circulating levels of PTX3. Further work is required to identify the mechanisms linking adiposity and PTX-3.

CT-derived pectoralis composition and incident pneumonia hospitalization using fully automated deep-learning algorithm: multi-ethnic study of atherosclerosis.

Pneumonia-related hospitalization may be associated with advanced skeletal muscle loss due to aging (i.e., sarcopenia) or chronic illnesses (i.e., cachexia). Early detection of muscle loss may now be feasible using deep-learning algorithms applied on conventional chest CT. To implement a fully automated deep-learning algorithm for pectoralis muscle measures from conventional chest CT and investigate longitudinal associations between these measures and incident pneumonia hospitalization according to Chronic Obstructive Pulmonary Disease (COPD) status. This analysis from the Multi-Ethnic Study of Atherosclerosis included participants with available chest CT examinations between 2010 and 2012. We implemented pectoralis muscle composition measures from a fully automated deep-learning algorithm (Mask R-CNN, built on the Faster Region Proposal Network (R-) Convolutional Neural Network (CNN) with an extension for mask identification) for two-dimensional segmentation. Associations between CT-derived measures and incident pneumonia hospitalizations were evaluated using Cox proportional hazards models adjusted for multiple confounders which include but are not limited to age, sex, race, smoking, BMI, physical activity, and forced-expiratory-volume-at-1s-to-functional-vital-capacity ratio. Stratification analyses were conducted based on baseline COPD status. This study included 2595 participants (51% female; median age: 68 (IQR: 61, 76)) CT examinations for whom we implemented deep learning-derived measures for longitudinal analyses. Eighty-six incident pneumonia hospitalizations occurred during a median 6.67-year follow-up. Overall, pectoralis muscle composition measures did not predict incident pneumonia. However, in fully-adjusted models, only among participants with COPD (N = 507), CT measures like extramyocellular fat index (hazard ratio: 1.98, 95% CI: 1.22, 3.21, p value: 0.02), were independently associated with incident pneumonia. Reliable deep learning-derived pectoralis muscle measures could predict incident pneumonia hospitalization only among participants with known COPD. Pectoralis muscle measures obtainable at zero additional cost or radiation exposure from any chest CT may have independent predictive value for clinical outcomes in chronic obstructive pulmonary disease patients. •Identification of independent and modifiable risk factors of pneumonia can have important clinical impact on patients with chronic obstructive pulmonary disease. •Opportunistic CT measures of adipose tissue within pectoralis muscles using deep-learning algorithms can be quickly obtainable at zero additional cost or radiation exposure. •Deep learning-derived pectoralis muscle measurements of intermuscular fat and its subcomponents are independently associated with subsequent incident pneumonia hospitalization.

The pK56M ICAM1 HFpEF risk variant and inflammatory biomarkers

IntroductionThe ICAM1 variant rs5491 (p.K56M) is common among Black individuals and has been associated with risk of heart failure with preserved ejection fraction (HFpEF). The pathways by which rs5491 leads to HFpEF are not known. MethodsAmong Black individuals within the Multi-Ethnic Study of Atherosclerosis, we evaluated associations of rs5491 with 3 inflammatory biomarkers (high-sensitivity C-reactive protein [hsCRP], interleukin-6 [IL-6], and tumor necrosis factor-α receptor 1 [TNFR-1]). ResultsAmong 1558 Black participants (mean age 62 ± 10 y, 47 % female), each additional rs5491 allele was associated with higher hsCRP after covariate adjustment (β: 0.15, SE: 0.07, P = 0.02). Each additional rs5491 allele was associated with higher TNFR-1 (β: 0.06, SE: 0.02, P = 0.02), but not IL-6 (β: 0.04, SE: 0.04, P = 0.29). The association between rs5491 and HFpEF remained significant after adjustment for hsCRP. ConclusionIn Black individuals, rs5491 (p.K56M) is associated with higher hsCRP and higher TNFR-1, but not IL-6.

Age at menarche and metabolic dysfunction-associated fatty liver disease: Evidence from a large population-based epidemiological study in Southwest China

BackgroundThe relationship between age at menarche and metabolic dysfunction-associated fatty liver disease remains largely not clear. The objective of this study was to examine the association between age at menarche (AAM) and metabolic dysfunction-associated fatty liver disease (MAFLD) in Chinese women and whether any observed associations were mediated by early adulthood adiposity. MethodsThe cross-sectional study included 46,873 Chinese women, aged 30–79 from baseline data of the China Multi-Ethnic Cohort study. Logistic regression models were used to evaluate the association between AAM and MAFLD. Mediation analysis was adopted to examine whether early adulthood adiposity (around 25 years) mediated the association between AAM and MAFLD. ResultsAAM was linearly and inversely associated with the risk of MAFLD (P for nonlinearity =0.743). In a multivariable-adjusted model, the odds ratios and 95% confidence interval (ORs (95% CI)) for MAFLD comparing menarche at <12, 12, 13, 15, 16, 17, ≥18 years to menarche at 14 years were 1.290 (1.082–1.537), 1.172 (1.068–1.285), 1.042 (0.960–1.131), 0.937 (0.861–1.020), 0.911(0.835–0.994), 0.868 (0.786–0.959), and 0.738 (0.670–0.814), respectively (P for trend <0.001). The 6.4% increased MAFLD risk was associated with each preceding year in AAM. The association between AAM and MAFLD was modified by age, ethnicity, and menopause. Early adulthood adiposity partially mediated this association. ConclusionThe findings of this study suggest that obesity prevention strategies are needed from young adulthood in women who undergo early menarche to reduce the risk of MAFLD.

Differences in left anterior descending coronary artery disease in participants without standard modifiable cardiovascular risk factors: the Multi-Ethnic Study of Atherosclerosis (MESA)

Abstract Background/Introduction The development of coronary artery disease (CAD) and myocardial infarction (MI) in the absence of standard modifiable cardiovascular risk factors (SMuRFs) has been highlighted as an unmet clinical need, accounting for between 10-27% of ST elevation MI and associated with higher 30-day mortality. One potential contributor to the early mortality is the higher observed rates of LAD territory culprit lesions in SMuRFless patients compared to those with at least one SMuRF. Purpose To characterise the differential proportion of LAD disease in an asymptomatic population with no prior cardiovascular disease (CVD) history, comparing individuals with CAD despite no SMuRFs to those with at least one SMuRF in the Multi-Ethnic Study of Atherosclerosis (MESA). Methods A MESA sub-cohort with a baseline coronary artery calcium score (CACS) &amp;gt;0 Agatston units (AU) was prepared. SMuRF status was assigned by self-report, treatment, or biomarkers for hypertension, hypercholesterolaemia, diabetes, and smoking. Proportion of disease in the left main, LAD, circumflex, and right coronary arteries were calculated by the vessel CACS divided by the total CACS. Standard descriptive statistics were used to characterise baseline differences. Group comparisons were completed with Wilcoxon rank sum for continuous measures and Chi-square for categorical. Results A CAC positive cohort of 3,417 participants was prepared from participants with SMuRF and CT imaging data, 14.3% of whom were considered SMuRFless. Total CACS was lower in SMuRFless participants compared to those with ≥1 SMuRF (58 AU vs. 102 AU, p&amp;lt;0.0001). However, proportion of calcium burden in the LAD was significantly higher in SMuRFless participants with CAD (74.2% vs. 56.8%, p&amp;lt;0.0001). When considering clinically significant CACS strata, proportion of LAD calcium burden was not significantly different between SMuRFless vs. ≥1 SMuRF participants with a total CACS of 1-99 AU (93.7% vs. 80.0%). However, statistically significant differences in LAD calcium proportion were observed between SMuRFless vs. ≥1 SMuRF participants with a total CACS 100-400 AU (69.5% vs. 57.8%) and CACS&amp;gt;400AU (45.2% vs. 40.5%). This difference remained significant during multivariate modelling, including consideration of absolute CACS. Conclusion Of patients with evidence of CAD (CACS&amp;gt;0) in the MESA cohort, SMuRFless status was associated with a higher proportion of LAD disease compared to participants with ≥1 SMuRF. This adds to previous observations in STEMI cohorts suggesting a heightened susceptibility to atherosclerosis development in the LAD territory against a lower traditional risk burden, and highlights the need for consideration of distinct biology and clinical significance in future studies.Fig. 1Fig. 2

IgE to common food allergens is associated with cardiovascular mortality in the National Health and Examination Survey and the Multi-Ethnic Study of Atherosclerosis

In individuals without symptomatic food allergy, food-specific IgE is considered clinically irrelevant. However, recent studies have suggested that galactose-α-1,3-galactose (alpha-gal) IgE is associated with cardiovascular (CV) disease. We sought to determine whether sensitization to common food allergens is associated with CV mortality. The association between IgE sensitization to foods and CV mortality ascertained to 2019 was examined in the National Health and Examination Survey (NHANES) 2005-2006 and the Wake Forest site of the Multi-Ethnic Study of Atherosclerosis (MESA) cohort; MESA enrolled adults without baseline clinical CV diseases between 2000 and 2002. Total and specific IgE was measured to cow's milk, egg, peanut, shrimp, and a panel of aeroallergens (NHANES), and to cow's milk, alpha-gal, peanut, dust mite, and timothy grass (MESA). Cox proportional hazard models were constructed, adjusting for sex, age, race/ethnicity, smoking, education, and asthma. A total of 4414 adults from NHANES (229 CV deaths) and 960 from MESA (56 CV deaths) were included. In NHANES, sensitization to at least 1 food was associated with higher CV mortality (hazard ratio [HR], 1.7 [95% confidence interval (CI), 1.2-2.4], P= .005). Milk sensitization was particularly associated (HR, 2.0 [95% CI, 1.1-3.8], P= .026), a finding replicated in MESA (HR, 3.8 [95% CI, 1.6-9.1], P= .003). Restricting analyses in NHANES to consumers of the relevant allergen strengthened food sensitization relationships, unmasking shrimp and peanut sensitization as additional risk factors for CV mortality. The finding that food sensitization is associated with increased risk of CV mortality challenges the current paradigm that sensitization without overt allergy is benign. Further research is needed to clarify mechanisms of this association.

A generalizable electrocardiogram-based artificial intelligence model for 10-year heart failure risk prediction

BackgroundHeart failure (HF) is a progressive condition with high global incidence. HF has two main subtypes: HF with preserved ejection fraction (HFpEF) and HF with reduced ejection fraction (HFrEF). There is an inherent need for simple yet effective ECG-based artificial intelligence (AI; ECG-AI) models which can predict HF risk early to allow for risk modification. ObjectiveThe main objectives were to validate HF risk prediction models using Multi-Ethnic Study of Atherosclerosis (MESA) data and assess performance of on HFpEF and HFrEF classification. MethodsECG-AI model was developed using raw 12-lead ECGs within a convolutional neural network (CNN). The clinical models from ARIC (ARIC-HF) and Framingham Heart Study (FHS-HF) used nine and eight variables, respectively. ECG-AI plus the clinical data within a Cox Proportional Hazard model (ECG-AI-Cox) and, for ECG-based comparison, a new Light Gradient Boosting Machine model using 288 ECG Characteristics (ECG-Chars) were developed. All the models were validated on MESA. The performances of these models were evaluated using the area under the receiver operating characteristic curve (ROC AUC) and compared using De-Long Test. ResultsECG-AI, ECG-Chars, and ECG-AI-Cox resulted in validation AUCs of 0.77, 0.73, and 0.84, respectively. ARIC-HF and FHS-HF yielded AUCs=0.76 and 0.74, respectively and CPH resulted in AUC=0.78. ECG-AI-Cox outperformed all other models. ECG-AI-Cox provided an AUC of 0.85 for HFrEF and 0.83 for HFpEF. ConclusionECG-AI utilizing ECGs provides better validated predictions when compared to HF risk calculators and the ECG feature model and also works well with HFpEF and HFrEF classification.

Carotid intima media thickness and cardiometabolic dysfunction: the Study of Women's Health Across the Nation.

Carotid artery intima media thickness (cIMT) and adventitial diameter (AD) are subclinical atherosclerosis indicators. Metabolic syndrome (MetS) and obesity are risk factors for atherosclerosis, but their combined impact on atherosclerosis risk is unknown. This study sought to examine the effect of the co-occurrence of MetS with obesity on cIMT and AD. The Study of Women's Health Across the Nation (SWAN) is a multi-center, multi-ethnic study. Carotid ultrasound assessments and concurrent physiologic measurements were undertaken between 2009 and 2013. This cross-sectional analysis included 1,433 women with body mass index ≥18.5 kg/m 2 and free of prevalent clinical cardiovascular disease. Multivariable linear regression models were used to relate maximum cIMT and AD (dependent variables) with obesity, MetS and their interaction. The average age was 60.1 years (standard deviation [SD], 2.7 y). The prevalence of obesity and MetS was 44% and 35%, respectively. Women with obesity had a 0.051 mm larger mean cIMT and women with MetS had a 0.057 mm larger cIMT versus women without the respective conditions (both P < 0.001). There was a statistically significant interaction between obesity and MetS ( P = 0.011); women with both had a model-adjusted predicted mean cIMT of 0.955 mm (95% confidence interval [CI], 0.897-1.013), higher than those with MetS alone (0.946 mm; 95% CI, 0.887-1.005), obesity alone (0.930 mm; 95% CI, 0.873-0.988), or neither condition (0.878 mm; 95% CI, 0.821-0.935). AD results were similar. Early detection and treatment of atherosclerotic changes may prevent significant disease. This study suggests there is a minimal impact of obesity on carotid artery thickness beyond MetS alone. All individuals with metabolic dysfunction, regardless of obesity status, should be considered at increased risk for atherosclerotic changes.

Abstract 17482: Association of Ankle-Brachial Index with Coronary Artery Calcium Scoring for Coronary Vascular Disease Risk Stratification

Introduction: The Ankle-Brachial Index (ABI) and coronary artery calcium scoring (CACS) are important tools for cardiovascular disease risk stratification. This study aimed to investigate the association between ABI index in combination with CACS for risk stratification of coronary vascular disease. Hypothesis: Given that both the ABI and the Coronary Artery Calcium Scoring are predictors of cardiovascular events, we hypothesize that an association between the Multiethnic Study on Subclinical Atherosclerosis Coronary Artery Calcium Scoring (MESA-CACS) and ABI exists. Methods: A total of 2655 patients (mean age 59±6 years, 72% male, and 87% white) who underwent CACS screening and ABI as part of a primary prevention screening program at the Cleveland Clinic between March 2016 and October 2021 were included in this study. ABI measurements were obtained by dividing the ankle systolic blood pressure by the brachial systolic blood pressure. The 10-year risk score based on MESA-CACS was calculated and categorized into 4 groups (&lt;1%, 1-4.99%, 5-9.99% and ≥10%). ABI was categorized into normal (1-1.4), non-compressible (≥ 1.4) and abnormal (&lt; 1). Results: Among the total of 2655 patients, 164 (6%) had a MESA-CAC score of &lt;1%, 1566 (59%) had a score of 1-4.99%, 501 (19%) had a score of 5-9.99%, and 424 (16%) had a score of ≥10%. The median ABI was 1.17 (range 0.79-1.77) on the left and 1.17 (0.50-1.78) on the right. Tabulating the patients based on their ABI index across the MESA-CACS categories showed no significant difference among the groups with a p-value of 0.830 (Fig 1), with the majority of the patients (&gt; 98% of patients) in all 4 groups having a normal ABI index of 1-1.4. Conclusion: The ABI did not demonstrate a significant association with the MESA-CACS, suggesting its potential use as a synergistic marker for coronary vascular disease risk stratification. Long-term cohort studies are needed to validate this result.

Abstract 13405: Cardiac Computed Tomography Variables for Prediction of New-Onset Heart Failure: Multi-Ethnic Study of Atherosclerosis

Introduction: Heart failure (HF) prediction models utilize clinical variables. The purpose of this study was to determine if addition of cardiac computed tomography (CT) variables could improve discrimination of HF prediction scores. Methods: Participants of the Multi-Ethnic Study of Atherosclerosis (MESA), a cohort between 45 to 84 years old and free of clinical cardiovascular disease at baseline, with complete clinical data and cardiac CT were included to evaluate new-onset HF during the follow-up period. Cardiac CT variables analyzed included left ventricular size index (LVSi) and calcification of the coronary arteries (CAC), aortic valve (AVC), mitral valve (MVC), and thoracic aorta (TAC). We evaluated if CT variables improved the Pooled Cohort equations to Prevent HF (PCP-HF) score (an internally and externally validated model of 33,010 patients predicting 10-year risk of new-onset HF). Results: Among 6,667 MESA participants (52.7% female), 426 events of new-onset HF occurred during the follow-up period. After the Cox model was adjusted for cardiac CT and PCP-HF score variables, CAC (HR 1.10; 95% CI 1.05-1.15; p&lt;0.001), AVC (HR 1.08; 95% CI 1.03-1.13; p=0.002), MVC (HR 1.06; 95% CI 1.01-1.12; p=0.015), and LVSi (HR 1.14; 95% CI 1.10-1.18; p&lt;0.001) were all significantly associated with new-onset HF. After using time-dependent ROC and stratifying by sex, AUC discrimination of new-onset HF with the PCP-HF score was improved by the inclusion of cardiac CT variables. In females, the AUC for 10-year risk of new-onset HF improved from 0.80 to 0.86 with the addition of cardiac CT variables. In males, the AUC for 10-year risk of new-onset HF improved from 0.77 to 0.84 with the addition of cardiac CT variables. Conclusions: In addition to traditional clinical risk factors, cardiac CT variables provide anatomic risk factors that improve the prediction of new-onset HF. Coronary and valvular calcification may signal poorly controlled clinical risk factors that are often associated with HF. LVSi may reflect left ventricular remodeling associated with HF. More studies are needed to further elucidate the relationship between cardiac CT anatomic risk factors and HF.

Abstract 18737: Coronary Artery Calcium for Allocation of Aspirin in Light of the 2022 USPSTF Guideline Recommendations: Results From the Multi-Ethnic Study of Atherosclerosis (MESA)

Introduction: The 2022 USPSTF guidelines issued a grade C recommendation for aspirin in patients aged 40-59 and recommended against its use in patients ≥ 60 years. We sought to assess if coronary artery calcium (CAC) would allow for personalized allocation of aspirin. Methods: Using MESA, aspirin naïve participants without high-risk bleeding features were identified. Participants were grouped by age into &lt;60, ≥ 60 and stratified by CAC and ASCVD risk. The number needed to treat (NNT 5 ) and number needed to harm (NNH 5 ) was assessed by applying a 11% relative risk reduction in CVD and a 44% relative risk increase in major bleeding (2022 USPSTF meta-analysis) to the observed CVD and bleeding events from MESA Results: Of 6,814 participants, 4,506 met inclusion criteria. 2,235 were &lt;60 years and 2,268 were ≥ 60 years. Figure-1 displays the interplay between baseline estimated ASCVD risk and CAC burden. In participants &lt;60 years of age, both CAC and ASCVD risk &gt;10% identified subgroups in whom the NNT was lower than the NNH: CAC 1-99 (NNT 5 -313, NNH 5 -517); CAC ≥100 (NNT 5 -152, NNH 5 -367); ASCVD risk &gt;10% (NNT 5 -155, NNH 5 -554) Fig-1. In those ≥ 60 years CAC ≥100 identified subgroups with a lower NNT 5 -128 than NNH 5 -168, while ASCVD risk stratification did not. 6% of adults &lt;60 years with &lt;10% ASCVD risk and 30.3% of adults ≥ 60 years who had CAC ≥100, and who demonstrated benefit from the use of Aspirin, would not qualify for the use of aspirin using only ASCVD risk as per the 2022 USPSTF guideline recommendations. Conclusions: CAC identifies subgroups of patients &lt;60 as well as those ≥ 60 years who would derive a net benefit from aspirin in primary prevention. A significant proportion of patients will be missed if only ASCVD risk scores are used for allocation of aspirin. CAC should be used to risk stratify individuals to personalize the allocation of aspirin

Abstract 13729: Subclavian Artery, Thoracic Aorta, and Coronary Artery Calcification, Age, Race/Ethnicity and Sex Distributions: The Multi-Ethnic Study of Atherosclerosis

Introduction: Upper extremity atherosclerosis is typically localized in the subclavian arteries, although it is less prevalent and less often symptomatic than in the lower extremities. Subclavian artery calcification (SAC) is a metabolic consequence of subclavian artery atherosclerosis. Hypothesis: SAC is different in gender and in different ethnicties. Aim: We studied SAC, in comparison with coronary artery calcification (CAC) and descending thoracic aorta calcification (TAC) in the MESA study population. Methods: We quantified SAC, TAC, and CAC in MESA participants who had a chest CT in 2016-18 (N = 2756, mean age 68.3±9.2 years, 48% male, 27% Black, 13% Chinese, 22% Hispanic, 38% White). We characterized the distributions of SAC and TAC, relative to CAC, and examined their associations with age, race/ethnicity, and sex. Race and sex differences were studied separately using logistic regression, adjusted for age, for each vascular bed. Results: Prevalence of any SAC was 62%, TAC 60%, and CAC 69%. Among those with any calcification, N, median and 90th percentiles were SAC: 1704, 76, 534; TAC: 1655, 152, 2011; CAC 1896, 135, 1118. Pairwise correlations [ln(Agatston score + 1)] were SAC - TAC: 0.61, SAC - CAC: 0.48, TAC - CAC: 0.48. Correlations with age were 0.46 for SAC, 0.58 for TAC and 0.40 for CAC. SAC and TAC prevalences were similar for males and females, but CAC was more common in males (Table, p&lt;0.05 for each race/ethnic group). Differences in prevalence varied by race, with small race differences for SAC, Chinese highest for TAC, and CAC highest for White (Table, groups with the same superscript within row do not differ significantly). Conclusion: In older adults, SAC and TAC occur almost as commonly as CAC. Calcification is moderately correlated between beds. The racial distribution is different for CAC than for SAC or TAC.

Abstract 12638: The Independent and Joint Associations of Obesity and Hypertension on Incident Heart Failure: A Pooled Cohort Analysis

Introduction: Obesity and hypertension (HTN) are major risk factors for incident heart failure (HF). Obesity is a major cause of HTN; thus, it is difficult to disentangle the impact of each risk factor on HF incidence. Further, it is not established what factors modify these relationships. Aims: To assess the associations of obesity and HTN as well as their potential synergistic effect on incident HF in large, diverse populations. Methods: We studied participants of 3 major epidemiologic cohorts: Atherosclerosis Risk in Communities Study, Multi-Ethnic Study of Atherosclerosis, and Jackson Heart Study. Participants were stratified into 4 groups: neither obesity nor HTN (reference), only obesity, only HTN, or both obesity and HTN. HTN was defined as ≥140 mmHg systolic or ≥90 mmHg diastolic blood pressure or use of antihypertensive medications. Obesity was defined as body mass index ≥30 Kg/m2. Cox proportional hazards models adjusted for age, sex, race, smoking, alcohol use, diabetes, coronary heart disease, and estimated glomerular filtration rate were used to assess the association of 4 groups with incident HF. Results: 18,351 participants (55.1% women and 31.4% black adults) followed over a mean 14 years were included in our analysis. Participants with only obesity (adjusted HR= 1.60; 95% CI 1.36 to 1.88) and only HTN (aHR 2.1; 95% CI 1.87, 2.37) demonstrated an increased risk of incident HF compared with participants with neither obesity nor HTN after adjustment for covariates. The greatest risk of incident HF was in participants with both obesity and HTN (aHR 2.97; 95% CI 2.63 to 3.36; Figure). HTN did not modify the effects of obesity or vice versa (p=0.193). Conclusions: Obesity and HTN are associated with incident HF and individuals with both have the highest risk of HF. This is an additive effect. These data emphasize the importance of blood pressure control and weight management in the prevention of HF.

Abstract 14460: Impact of Carotid Artery Plaque Detection on Cardiovascular Disease Risk and the Importance of Age in the Multi-Ethnic Study of Atherosclerosis

Background Carotid artery plaque (CAP) predicts future atherosclerotic cardiovascular disease (ASCVD) events; however, its prevalence increases with age and differs by sex and race/ethnicity, so how its presence influences ASCVD risk in clinical settings is uncertain. Hypothesis: We hypothesized that the impact of CAP detection on relative ASCVD risk declines with age. Methods: CAP was detected and quantified by ultrasound in adults free of clinical ASCVD at Exam 1 of the Multi-Ethnic Study of Atherosclerosis who were followed longitudinally for coronary death, myocardial infarction, angina, stroke, and ASCVD death. Cox proportional hazards models investigated associations of CAP and incident ASCVD events in subsets defined by age categories. These models were adjusted for confounders and an age*CAP interaction. Results: The 5198 participants were a mean (standard deviation) 61.8 (10.1) years old (53% female, 27% Black, 22% Hispanic, 12% Chinese) and experienced 880 ASCVD events over 14.3 (5.7) years. Prevalence of CAP and the age at which 50% of participants had detectable CAP differed by age, sex, and race/ethnicity, ranging from 13% in Chinese women &lt;50 years old to 97% in Non-Hispanic White men 80-84 years old (Figure). CAP presence independently predicted ASCVD events (hazard ratio, 1.35, [95% confidence interval 1.17-1.57], p&lt;0.001), however the strength of this association declined with increasing age (p interaction =0.029, Figure). CAP detection in younger individuals conferred higher relative ASCVD risk than in older participants who had higher absolute risk, regardless of CAP. Conclusions: Prevalence of CAP varies by sex and race/ethnicity. CAP becomes very common with increasing age and its relative association with future ASCVD events declines. These data can assist in guiding ASCVD risk assessment in younger adults and provide perspective for incidentally detected CAP on clinical imaging studies in older adults who have higher absolute ASCVD risk.

Abstract 14697: Developing a Machine Learning Algorithm Using Routine Clinical Data for the Detection of Hard Cardiovascular Disease Events

Introduction: Machine learning (ML) may enhance prediction of hard cardiovascular disease (CVD) events (e.g., myocardial infarction, stroke, cardiac arrest). Developing ML algorithms using routinely-collected clinical data offers a practical method to improve CVD event prediction. This study aimed to train ML algorithms to predict 10-year risk of hard CVD events using common clinical variables and compare performance to pooled cohort equations (PCEs). Hypothesis: ML algorithms trained on data from the Multi-Ethnic Study of Atherosclerosis (MESA) and Atherosclerotic Risk in Communities (ARIC) would outperform PCEs in predicting 10-year risk of hard CVD events when tested on a PCE-valid subset. Methods: Data from MESA and ARIC were obtained. Pooled baseline data were used for algorithm training. Patients with prior CVD or lacking 10-year follow-up were excluded. The primary endpoint was hard CVD event within 10 years of baseline. Variables of interest included patient history, anthropometry, and serum lipid levels. Cohort data was split into training (75%) and test (25%) sets. Models trained were neural network, gradient boost machine, k-nearest neighbors, and an ensemble model. Results: In total, 17,775 patients (5,703 MESA, 12,072 ARIC) with mean age of 56.2 ( + 8.2) years and 54.6% female were included. During the 10-year follow-up period, 1,344 (7.6%) had a first-time hard CVD event. The ensemble model had the highest AUC on the full test set (0.783 [95% CI 0.76-0.81]), followed by the gradient boosting machine (0.781 [0.76-0.81], neural network (0.780 [0.76-0.81]), and k-nearest neighbors (0.737 [0.71-0.76]). The gradient boosting machine demonstrated highest sensitivity (76.0%), while the ensemble had highest specificity (78.6%). In the PCE-valid subset, the ensemble model had the highest AUC (0.780 [0.75-0.81]) and outperformed PCE (AUC = 0.757 [0.73-0.78], p = 0.001), as did the gradient boosting machine (AUC = 0.779 [0.75-0.81], p = 0.003]) and neural network (AUC = 0.776 [0.75-0.80], p = 0.007). Conclusions: The ensemble, gradient boosting machine, and neural network improved prediction of hard CVD events over PCEs within a PCE-valid subset despite training on a broader cohort. Future studies should validate results on an external dataset.

Abstract 16389: Inflammatory Chronic Kidney Disease is Associated With Incident Heart Failure: A Pooled Cohort Analysis

Introduction: High sensitivity C-reactive protein (hsCRP), a biomarker of inflammation, can risk stratify patients for cardiovascular disease including heart failure (HF). It is not well established if inflammation increases the risk of HF in patients with chronic kidney disease (CKD) and if hsCRP can be utilized to risk stratify patients with CKD for their risk of incident HF. Aims: We assessed the role of hsCRP in risk stratifying patients with CKD for incident HF. Methods: We studied participants of 3 major epidemiologic cohorts: Atherosclerosis Risk in Communities Study, Multi-Ethnic Study of Atherosclerosis, and Jackson Heart Study. High inflammation was classified as hsCRP &gt;3 mg/L and CKD was classified as estimated glomerular filtration rate &lt;60 mL/min/1.73m2. Participants were stratified into 3 groups: neither CKD nor high inflammation, CKD without high inflammation, or CKD with high inflammation. Cox hazards models adjusted for age, sex, race, body mass index, smoking, alcohol use, diabetes, hypertension, and coronary disease were utilized for analysis. Results: 8,692 participants with 460 incident HF events followed over a median 8.32 years were included in our analysis. Participants with CKD without high inflammation (HR = 1.45; 95% CI 1.14 to 1.85) and CKD with high inflammation (2.05; 1.61 to 2.61) were associated with an increased risk of incident HF compared with participants without CKD or high inflammation (referent, Figure). Furthermore, participants with CKD and a hsCRP &gt;3 mg/L had an increased risk of incident HF (1.41; 1.06 to 1.88) compared with participants with CKD but hsCRP ≤3 mg/L (referent). Conclusions: Patients with CKD and high levels of inflammation carry a 2-fold increased risk of incident HF. In patients with CKD, hsCRP levels &gt;3mg/L may be useful to risk stratify patients with increased risk of developing HF.

Abstract 13805: Associations of Subclavian Artery Calcification With Future Cardiovascular Events and Death: The Multi-Ethnic Study of Atherosclerosis

Introduction: Atherosclerosis affects the upper limbs and may be a marker of diffuse atherosclerosis and increased risk for cardiovascular events. Hypothesis: Subclavian artery calcification (SAC) and thoracic aorta calcification (TAC) obtained by chest CT provide additional information about future CV events and death beyond coronary artery calcification (CAC) and CV risk factors. Methods: Chest CT scans of MESA participants (N = 2519) at exam 5 (2016-18) were interrogated for the presence and extent of SAC, TAC, and CAC. Proportional hazards regression was conducted for associations with incident total CVD events (n = 209), adjusted for age, race/ethnicity, and sex, then SAC, TAC, and CAC adjusted for demographics and each other, and finally adding adjustment for CV risk factors. Results: The mean age of the sample was 68.3±9.2 years; 48% were male, 27% Black, 13% Chinese, 22% Hispanic, 38% White. The prevalence of any calcification was 62% for SAC, while being 60 and 69% for TAC and CAC, respectively. Among those with any calcification, N and median Agatston score were SAC: 1704, 76; TAC: 1655, 152; CAC 1896, 135. CAC predicted future events in all models (Table). After adjustment for demographics, the ln(Agatston score +1) in each calcified vascular bed predicted future CVD. Mean±SD of the predictors were ln(SAC+1): 2.4±2.4, ln(TAC+1) 2.7±2.9 and ln(CAC+1) 3.1±2.6. SAC, but not TAC, was still predictive after mutual adjustment. SAC was marginally predictive (HR: 1.08, p = 0.06) when CV risk factors were also included in the model. Conclusions: Calcification was common in all 3 vascular beds, but there is no evidence that predictions using SAC and TAC are independent of CAC. We speculate that although the tendency to calcify is similar across vascular beds, CAC is most relevant for incident CVD.

Abstract 12706: The Impact of Social Determinants of Health and Lifestyle Factors on the Association of Lipoprotein(a) With Cardiovascular Outcomes in Two US Cohorts

Introduction: In European cohorts, a higher Mediterranean diet or Life’s Simple 7 (LS7) score abolished or attenuated the risk associated with increasing Lipoprotein(a) [Lp(a)] on cardiovascular (CV) outcomes. This is unstudied in US cohorts. Further, the impact of social determinants of health (SDOH) on the association of Lp(a) with CV outcomes is unstudied. Hypothesis: We hypothesized that SDOH and LS7 scores would attenuate the association of Lp(a) with CV outcomes in US cohorts. Methods: We used data from the Atherosclerosis Risk in Communities (ARIC) and the Multi-Ethnic Study of Atherosclerosis (MESA) studies. Outcomes were time until first fatal or nonfatal myocardial infarction (MI) or stroke. We performed multivariable cox proportional hazard analyses with covariates of age, gender, race and ethnicity, SDOH score, and LS7 score. We identified five SDOH in ARIC and 11 in MESA. The Lp(a) protein-mass from ARIC was tripled to be similar to the Lp(a) total-mass measurement in MESA. Results: ARIC included 15,072 individuals (median Lp(a) = 17.3 mg/dL (IQR 7.5, 40.6)) with 16.2 years average follow up. MESA included 6,822 individuals (Median Lp(a) = 18.3 (IQR 6.9, 43.8)) with 12.3 year average follow up. In age, gender, and race and ethnicity adjusted analyses Lp(a) was associated with MI in ARIC (HR 1.10, p&lt;0.001) and MESA (HR 1.09, p&lt;0.001). Lp(a) was also associated with stroke in ARIC (HR 1.08, p&lt;0.001) but not MESA (HR 0.97, p=0.50). With SDOH and LS7 added to the model, association of Lp(a) with CV outcomes remained similar in ARIC and MESA. SDOH correlated positively and LS7 correlated negatively with MI and stroke. In MESA there was an interaction between LS7 score and Lp(a) for MI. Conclusions: Lp(a) remained similarly associated with CV outcomes in US cohorts after accounting for social and lifestyle factors, although LS7 score may have a small modifying effect on the association between Lp(a) and MI. SDOH and LS7 are strong predictors for CV outcomes.

Abstract 12716: Left Ventricular Hypertrophy Phenotype to Predict Incident Atrial Fibrillation: The Multi-Ethnic Study of Atherosclerosis

Introduction: Left ventricular hypertrophy (LVH) has been shown to be associated with the occurrence of atrial fibrillation (AF). However, due to the heterogeneous nature of LVH, the predictive value of the LVH phenotype for incident AF remains uncertain. Methods: We utilized data from the Multi-Ethnic Study of Atherosclerosis (MESA). In this study, participants who underwent cardiac magnetic resonance imaging examination were included for analysis. LVH was defined as LV mass index ≥ 85.3 g/m 2 for females and ≥ 107.8 g/m 2 for males. Isolated LVH was determined as LVH without elevated cardiac biomarkers, and malignant LVH was determined as LVH with elevated hs-TnT level of ≥14 ng/L, or/and NT Pro-BNP level of ≥125 pg/mL. Results: A total of 4983 participants were included in the final analysis, with a mean age of 61.5 years. 272 (5.6%) had isolated LVH, and 222 (4.5%) had malignant LVH. During a median follow-up of 8.5 years, 272 cases of AF were observed. Compared to participants without LVH or elevated cardiac biomarkers, those with isolated LVH (HR, 1.82; 95% CI, 1.03-3.20) and malignant LVH (HR, 4.13; 95% CI, 2.77-6.16) had higher AF risks after adjusted for age, sex, race, body mass index, systolic blood pressure, antihypertensive medication use, smoking status, and diabetes. Malignant LVH carried a 1.5-fold increased risk (HR: 2.48, 95% CI: 1.30-4.73) of AF compared to isolated LVH (Panel A). Kaplan-Meier cumulative AF incidence curves showed that malignant LVH had a significantly higher incidence of AF than those in other phenotypes (Panel B). Including the LVH phenotype in the CHARGE-AF model improved model discrimination (areas under the curves[AUC] increase: 0.03, p&lt;0.001) (Panel C). Conclusions: The risks of AF incidence varied across LVH phenotypes, showing that malignant LVH carried a higher risk compared to other LVH phenotypes. LVH phenotype could provide additional predictive power for AF occurrence above the CHARGE-AF model.