The use of coronary computed tomography angiography (CCTA) to predict coronary artery plaque, with a focus on new, high-risk, and obstructive plaques

To develop and validate a predictive model based on CCTA-derived fat attenuation index (FAI) for the formation of new, high-risk, and obstructive plaques.

This research retrospectively included 160 patients with cardiovascular risk factors for model development and 60 patients for external verification. All patients underwent at least two CCTA examinations within a 5-year period, and initial results were normal. Newly formed plaques, along with their vulnerability and luminal stenosis was assessed. The FAI was quantitatively evaluated on the three major coronary arteries with the initial CCTA.

Among 160 participants, newly formed plaques were identified in 60 patients. FAI had a certain value in predicting new (area under the curve (AUC) 0.723 [95% CI: 0.645, 0.802], with a cut-off value of -80.5 HU, sensitivity of 73.3% and specificity of 64.0%), high-risk (AUC 0.822 [95% CI: 0.715, 0.929], with a cut-off value of -78.5 HU, sensitivity of 90.9% and specificity of 65.3%), and obstructive (AUC 0.823 [95% CI: 0.720, 0.926], with a cut-off value of -77.5 HU, sensitivity of 90.0% and specificity of 74.0%) plaques and exhibited an incremental value compared with conventional clinical risk factors (0.817 vs 0.749; 0.862 vs. 0.700; 0.876 vs 0.740, all P < 0.001).

FAI and cardiovascular risk factors aided in predicting the probability of coronary plaque formation (FAI: -80.5 HU), especially high-risk (FAI: -78.5 HU) and obstructive (FAI: -77.5 HU) plaques.