The role of pericoronary adipose tissue inflammation in atherosclerotic plaque progression and vulnerability: insights from three different CCT-based studies

Background Inflammation of pericoronary adipose tissue (PCAT) is a critical driver of atherosclerotic plaque progression and destabilization. Although cardiac computed tomography (CCT) provides detailed anatomical evaluation of coronary artery disease, traditional CCT-derived markers, including stenosis severity and coronary artery calcium score (CACs), fail to fully capture plaque biological activity. Purpose The aim of these studies was to investigate the clinical significance of PCAT inflammation using CCT-based artificial intelligence (AI)–supported radiomic analysis and quantitative assessment of the fat attenuation index (FAI) across different pathological and therapeutic settings. Methods A total of 668 patients presenting with chest pain were enrolled. All underwent CCT for evaluation of CACs, plaque morphology, and PCAT-derived FAI, with quantitative analysis performed on more than 2,000 coronary segments. Three substudies were conducted: post–COVID-19 inflammation (n=255), CACs-based vulnerability analysis (n=336), and longitudinal assessment of intensive statin therapy (n=77). Results Elevated FAI values were consistently associated with high-risk plaque (HRP) morphology and adverse coronary phenotypes, supporting FAI as a marker of biologically active atherosclerosis. Post–COVID-19 patients showed significantly higher FAI-Score centiles across all major coronary territories compared with controls (overall p<0.05), indicating persistent inflammation despite non-obstructive disease. The RCA exhibited greater inflammatory burden than the LCA (p<0.01). In the CACs substudy, HRP features were present even in patients with CACs=0 (10.87%), and nearly 9% were classified as high risk by AI-based profiling. Although CACs moderately correlated with disease burden and stenosis severity (p<0.0001), its association with inflammation was weak (r²=0.08), highlighting partial dissociation between calcification and inflammatory activity. Lipid-rich vulnerable plaques with elevated FAI were identified in symptomatic CACs=0 patients, including cases requiring revascularization. In the longitudinal statin cohort, intensive therapy significantly reduced FAI across all vessels (p<0.005) and was accompanied by favorable plaque remodeling, with increased calcified/fibrotic components and reduced lipid-rich volume (p<0.05). FAI correlated positively with non-calcified and lipid-rich plaque volumes (p<0.001). Conclusions AI-enhanced quantification of PCAT inflammation using FAI independently identifies biologically active, high-risk coronary plaques across diverse clinical settings. The detection of vulnerable plaques in CACs=0 patients, persistent post–COVID-19 inflammation, and the measurable anti-inflammatory effects of statin therapy collectively underscore the limitations of purely anatomical assessment and support a transition toward inflammation-guided cardiovascular risk stratification and personalized preventive cardiology.