However, the plaque quantification model demonstrated a 15.3 percent higher AUC, 8.3 percent higher sensitivity, 14.4 percent higher specificity and 12.1 percent higher accuracy in comparison to the clinical model.
“Traditional risk factors indicate upstream susceptibility to atherosclerosis, but do not directly quantify disease burden. In contrast, quantitative carotid plaque volume provides an imaging-based measure of systemic atherosclerotic burden and reflects pathological features shared with coronary disease, such as lipid deposition, fibrosis, and calcification. These characteristics may underlie the superior diagnostic value of plaque metrics over clinical variables for identifying OCAD,” noted lead study author Lusiying Xu, MD, who is affiliated with the School of Medical Imaging at Xuzhou Medical University in Xuzhou, China, and colleagues.
Emphasizing the potential triage value with adjunctive use of the plaque quantification model, the study authors pointed out that the AI-powered quantification model accurately identified 74.1 percent of OCAD cases and correctly indicated the 56 percent of patients were at low risk for OCAD.
Three Key Takeaways
- Routine carotid CTA can flag likely coronary disease. Automated quantification of carotid plaque on existing carotid CTA detected obstructive CAD (> 50 percent stenosis on CCTA) better than a clinical risk factor model. In external validation, the AUC was 81.3 percent vs. a 66 percent AUC for the clinical model. Adding clinical variables to the plaque model provided no meaningful gain in detection.
- Plaque composition matters, especially calcified and fibrotic volume. Greater calcified carotid plaque volume was linked to more than double the odds of obstructive CAD. Greater fibrotic plaque volume was linked to 37 percent higher odds.
- Consider the possible adjunctive triage benefits. The integrated model identified 74.1 percent of obstructive CAD cases and correctly classified 56 percent of patients as low risk. That makes automated carotid plaque quantification a practical aid for deciding which patients undergoing carotid CTA may warrant further coronary evaluation.
What the Research Reveals About the Link Between Calcified Carotid Plaque Volume and Obstructive CAD
Multivariable analysis revealed that greater calcified plaque volume was associated with over double the likelihood for OCAD and greater fibrotic plaque volume was linked to a 37 percent higher likelihood of OCAD, according to the researchers.
“These findings highlight the potential of using automated quantitative analysis of existing carotid CTA images as an adjunctive risk-stratification tool for coronary heart disease risk assessment,” added Xu and colleagues.
(Editor’s note: For related content, see “FDA Clears CTA-Based AI Software for Enhanced Visualization of Head and Neck Arteries,” “Comparative AI Study Shows Merits of RapidAI LVO Software in Stroke Detection” and “Why Plaque Burden is Critical to Assessing Cardiovascular Risk: An Interview with Ibrahim Danad, MD, PhD.”)
In regard to study limitations, the authors acknowledged the retrospective design and conceded that a cohort of exclusively Chinese participants may thwart application of the study results to a broader population. The researchers also noted that the evaluated model in the study predicts anatomically defined stenosis as opposed to clinically significant ischemic disease.