Inflammatory risk quantification in systemic inflammatory disease using coronary CT imaging

Abstract Background Immune-mediated Inflammatory Diseases (IMIDs) are associated with increased cardiovascular risk. Coronary inflammation is a key component of atherosclerosis. Tracking spatial attenuation changes in perivascular adipose tissue driven by inflammatory signals secreted from the adjacent artery has been introduced as a new way to non-invasively quantify inflammation using routine coronary computed tomography angiography (CCTA). Purpose To test the ability of coronary inflammation imaging among individuals with IMIDs to predict adverse cardiovascular outcomes. Methods IMIDs patients clinically referred for CCTA at two academic medical centers were included. CAD severity was determined using natural language processing from electronic health records and was categorized using the CAD-RADS 2.0 classification: no plaque or stenosis (CAD-RADS 0), non-obstructive CAD (CAD-RADS 1-2, 1-49% stenosis), and obstructive CAD (CAD-RADS 3-5, ≥50% stenosis). Coronary inflammation was measured using the perivascular Fat Attenuation Index (FAI) Score, an established quantitative metric of coronary inflammation (adjusted for age, sex, and technical parameters) in each major coronary artery and averaged to create a single FAI Score per scan. The primary endpoint was a composite of death from cardiovascular causes, myocardial infarction, or hospitalization for heart failure. The secondary endpoint was a composite of death from cardiovascular causes, myocardial infarction, or stroke. Cox proportional-hazards models comparing patients with low (1st-74th percentile) versus high (75th-100th percentile) FAI Scores and adjusted for the number of traditional cardiovascular risk factors (hypertension, hyperlipidemia, diabetes mellitus, current smoking status) and the extent of CAD using CAD-RADS were used. Results 909 patients with mean age of 62 years (interquartile range (IQR), 52 to 70) were included. 56.7% were women and 26.4% had 3 or more risk factors. Over a median follow-up of 5.5 years the primary endpoint occurred in 69 patients (26%) in the high FAI Score group as compared with 87 patients (13.5%) in the low FAI Score (adjusted hazard ratio, 1.47; 95% confidence interval CI, 1.06-2.04; p=0.02; panel a). The hazard ratio for the secondary endpoint was 1.46 (95% CI, 1.04-2.04, panel b). In exploratory subgroup analyses, the association of FAI with the primary outcome was higher among younger (age 50) individuals. High FAI Score maintained its association with the primary outcome among patients with 50% stenosis. Among IMID patients receiving treatment with either systemic steroids, disease-modifying anti-rheumatic drug (DMARD) biologics or JAK inhibitors the association of FAI Score with the primary outcome persisted (panel c). Conclusions Coronary inflammation from routine CCTA captures cardiovascular inflammatory risk in IMID patients, including the young and those without obstructive CAD, and could serve as a novel risk prediction tool.