Key result
Adding CCTA plaque features to stenosis improves ischemia prediction in LAD, gaining ~0.14 AUC.
Why the study?
CCTA-derived plaque characteristics are associated with myocardial ischemia, but whether these relationships with FFR-defined ischemia and their incremental value beyond stenosis severity differ across coronary vessels remained unclear.
Does adding vessel-specific quantitative CCTA plaque characteristics improve the prediction of FFR-defined myocardial ischemia beyond clinical variables and severe stenosis in patients undergoing CCTA and invasive FFR?
Cohort (n=204)
Blinded outcome assessment (radiologists blinded to clinical outcomes)
No
Does adding vessel-specific quantitative CCTA plaque characteristics improve the prediction of FFR-defined myocardial ischemia beyond clinical variables and severe stenosis in patients undergoing CCTA and invasive FFR?
Quantitative CCTA plaque characteristics demonstrate vessel-specific associations with FFR-defined myocardial ischemia and provide incremental predictive value beyond stenosis severity in LAD and RCA vessels.
Answers are grounded in this paper and cite their sources.
May enhance CCTA ischemia prediction in LAD/RCA vessels; extends observational associations but leaves open prospective validation.
Objectives : Coronary CT angiography (CCTA)-derived plaque characteristics are associated with myocardial ischemia, but whether these relationships differ across coronary vessels remains unclear. We aimed to investigate vessel-specific associations between quantitative plaque characteristics and fractional flow reserve (FFR)-defined myocardial ischemia nd assess their incremental value beyond stenosis severity. Materials and Methods : This retrospective single-center study included 204 patients comprising 290 coronary vessels (161 LAD, 56 LCX, and 73 RCA) who underwent CCTA followed by invasive coronary angiography with FFR measurement. Quantitative plaque analysis was performed using semi-automated software. Vessel-specific feature selection was conducted using LASSO or Boruta combined with XGBoost within a five-fold cross-validation framework. Bootstrap resampling was used to assess model robustness. Incremental value was evaluated by comparing a baseline model incorporating clinical variables and severe stenosis (≥70%) with an extended model including selected plaque characteristics. Results : Ischemic vessels exhibited longer lesion length, smaller minimal lumen area, greater stenosis severity, and higher plaque burden than non-ischemic vessels. Bootstrap validation confirmed vessel-specific plaque signatures associated with ischemia. Plaque-based models achieved good discrimination for FFR-defined ischemia, with AUCs of 0.91 (95% CI: 0.87–0.97), 0.94 (0.86–1.00), and 0.97 (0.89–1.00) for LAD, LCX, and RCA, respectively. Addition of plaque characteristics significantly improved discrimination and reclassification in LAD (AUC: 0.613 vs. 0.755; NRI=0.736; IDI=0.163) and RCA vessels (AUC: 0.489 vs. 0.767; NRI=0.955; IDI=0.276), but not in LCX vessels. Conclusion : Quantitative CCTA plaque characteristics show vessel-specific associations with FFR-defined myocardial ischemia and provide incremental predictive value beyond stenosis severity, particularly in LAD and RCA vessels.
No takes yet. Share an insight, caveat, or question.
Li et al. (2026) conducted a cohort in Suspected or known coronary artery disease and myocardial ischemia (n=204). Quantitative CCTA plaque characteristics model (Model 2) vs. Baseline model with clinical variables and severe stenosis ≥70% (Model 1) was evaluated on Vessel-level functional myocardial ischemia (defined as invasive hyperemic FFR ≤ 0.80) in the LAD (ΔAUC 0.142, 95% CI 0.061-0.217, p=0.002). Adding CCTA plaque features to stenosis severity improved ischemia prediction in LAD (AUC 0.613 to 0.755, P=0.002) and RCA (AUC 0.489 to 0.767, P<0.001), but not LCX.
Synapse has enriched 10 related papers. Consider them for comparative context: