Abstract
Background: Perivascular fat attenuation index (FAI) derived from coronary CT angiography (CCTA) has emerged as a quantitative biomarker of vascular inflammation, with potential to improve risk stratification in coronary artery disease (CAD) patients. This study aimed to evaluate plaque characteristics of coronary atherosclerotic lesions in patients with high (≥-70.1 HU) or low FAI of pericoronary adipose tissue. Methods: In a retrospective analysis, patients with suspected or confirmed CAD who underwent coronary CTA were included. Coronary lesions were classified into two groups based on their perivascular inflammation as assessed by CCTA: high perivascular FAI phenotype (≥-70.1 HU) versus low FAI phenotype (<-70.1 HU). Both groups were compared with respect to various patient- and lesion-specific characteristics. Results: A total of 247 coronary lesions were analyzed in this study. Of these, 36 (14.6%) lesions were associated with high perivascular inflammation (high FAI phenotpye) and 211 (85.4%) were associated with low perivascular inflammation (low FAI phenotype). Lesions with a high FAI phenotype demonstrated a significantly higher amount of non-calcified plaque volume (NCPV) compared to lesions with a low FAI phenotype [(111.8 mm(3) (69.4-184.2) versus 87.7 mm(3) (44.6-143.0), p < 0.003]. NCPV emerged as a consistent and significant predictor of fat attenuation positive plaque in both univariate (OR 1.030 [95% CI, 1.010-1.050], p = 0.003); and multivariate logistic regression analyses (OR 1.028 [95% CI, 1.008-1.050]. p = 0.007). Additionally, lesions with a high FAI phenotype less frequently exhibited homogeneous calcification than their low FAI phenotype counterparts (25% versus 46.9%, p = 0.014). Conclusions: Coronary lesions associated with a high FAI phenotype on coronary CCTA consist predominantly of non-calcified plaques. Conversely, lesions characterized by a low perivascular FAI phenotype are primarily calcified and seem to be more homogeneous by visual assessment. Further prospective studies are warranted to validate these associations and explore the underlying pathophysiological mechanisms.