Epicardial Adipose Tissue Analysis Identifies Increased Non-Calcified Plaque Burden in Non-Obese Individuals with Low CAC Scores: An AI-CVD Study within the Miami Heart Study
K Atlas, C Zhang, A Reeves, A Azimi, R Mirjalili, M MozafaryBazargany, A Ghaffari Jolfayi, A Hashemi, H Ghoshouni, H Mohammadi, H Zarei, Z Fayad, D Yankelevitz, N Wong, M NaghaviAbstract
Background
Coronary artery calcium (CAC) scoring is now widely accepted as the method of choice for noninvasive detection of subclinical atherosclerosis. However, a very small subset of CAC zero individuals experiences a coronary event presumably related to complications of non-calcified plaques (NCP). Epicardial adipose tissue (EAT) reflects cardiometabolic risk and may contribute to coronary artery disease.
Purpose
To evaluate the association between EAT volume and non-calcified plaque burden across CAC categories, and to determine whether EAT identifies a high-risk plaque phenotype among individuals with low CAC and non-obese BMI.
Methods
We analyzed 1,259 participants from the Miami Heart Study with paired CAC and coronary CT angiography (CCTA) scans. EAT volume was automatically quantified using FDA cleared AI-CVD software from CAC images, and NCP volumes were derived from CCTA. Multivariable linear regression models assessed the association between standardized EAT and log-transformed NCP, adjusting for age and CAC, with interaction testing across traditional CAC categories (0, 1–99, 100–399, ≥400). Additional analyses were performed after adjustment for BMI and within a restricted cohort of individuals with BMI <29 kg/m². A discordant phenotype analysis compared NCP burden among participants with low CAC (0–99) across EAT quartiles.
Results
In the overall cohort, higher EAT was associated with greater NCP after adjustment for age and CAC (P=0.045), with no significant interaction by CAC category (P-interaction=0.62). However, the association attenuated after BMI adjustment. In contrast, among individuals with BMI <29 kg/m² (n=758), higher EAT was associated with nearly a two-fold higher NCP per standard deviation increase (93% increase; P=0.072). Within the low-CAC subgroup (0–99), participants in the highest EAT quartile demonstrated a three-fold higher median NCP burden compared with the lowest quartile (34.5 vs 11.9 mm³, P<0.001), corresponding to a 6-fold higher adjusted NCP volume (P=0.002).
Conclusion
EAT volume is associated with greater non-calcified coronary plaque burden after adjustment for age and CAC score, with the strongest associations observed among non-obese individuals. EAT may identify a metabolically adverse phenotype in patients with low CAC and normal-to-overweight BMI, suggesting that routine CAC scans may contain additional cardiometabolic information beyond calcium scoring alone. Investigations are warranted into whether adding EAT volume to CAC and CCTA reports can improve patient care.case example