Body composition and atherosclerotic plaque activity in patients with established coronary artery disease

Abstract 18F-sodium fluoride (18F-NaF) positron emission tomography (PET) has shown hope in refining risk stratification in patients with established coronary artery disease (CAD). Computed tomography is increasingly being employed for assessments of body composition, with this novel application providing additional prognostic value across a wide range of CAD manifestations. To date we lack data on the influence of body composition on outcomes in the challenging population of patients with established CAD. Purpose We aimed to elucidate the complex interplay between body composition, plaque activity and outcomes. Methods Patients with known CAD (with a history of either myocardial infarction or coronary revascularization) underwent 18F-NaF PET. Coronary 18F-NaF uptake was determined by the coronary microcalcification activity (CMA). Body composition quantification was performed with deep learning on low dose computed tomography acquired for attenuation correction purposes. Tissue types were differentiated using Hounsfield unit (HU) ranges. Adipose tissue within the intrathoracic space was considered visceral adipose tissue, unless segmented as epicardial adipose tissue. Four tissues (subcutaneous adipose tissue, skeletal muscle, intramuscular adipose tissue (IMAT), and bone) were segmented in the extrathoracic space. Non-fatal myocardial infarction and cardiac death served as the composite primary endpoint. Results The study population included 759 patients (age 64±8.0, 117 [15%] women) with a history of coronary revascularization and a high disease burden with a median coronary artery calcium score of 190 [52-571] Agatston units. On 18F-NaF PET overall 461 (61%) patients showed 18F-NaF coronary uptake (CMA>0). The median CMA was 0.27 [0.0-2.32] with 288 (38%) patients presenting with 1 or 2 vessels with CMA>0 and 173 (23%) with triple-vessel 18F-NaF increased uptake. Over a median of 4.5 [4.0 to 6.0] years and 3573 patient-years of follow-up 64 patients experienced the composite primary endpoint. Among all the baseline clinical and imaging characteristics only IMAT volume (250.0±121.7 vs 218.6±98.2, p=0.015) and CMA (1.6 [0.9-2.5] vs 0.6 [0.0-1.6], p=0.04) were significantly higher in those with an event. Indeed on cox proportional hazard modelling IMAT volume >221.1 (HR 1.7; 95% CI 1.0-2.8, p=0.04) and CMA greater than 0 (HR 2.4; 95% CI 1.3-4.3, p=0.004) emerged as predictors of the composite primary outcome. These associations were independent of age and the coronary artery calcium score and persisted on multivariable cox regression modelling - HR 1.6; 95% CI 1.0-2.8, p=0.049 and HR 2.4; 95% CI 1.3-4.5, p=0.004; for IMAT volume and CMA, respectively). Conclusions While 18F-NaF coronary PET remains the strongest predictor of the composite endpoint of myocardial infarction or cardiac death, body composition analysis has the potential to enhance risk stratification in the challenging cohort of patients with established CAD.