Quantitative plaque characteristics and pericoronary fat attenuation index enhance risk prediction of unstable angina in nonobstructive lesions.

AIM: The role of quantitative plaque characterization and pericoronary fat attenuation index (FAI) in nonobstructive lesions is uncertain. Hence, this study aimed to investigate artificial intelligence (AI)-based plaque characterization and pericoronary FAI in patients with nonobstructive lesions to enhance risk prediction of unstable angina.

MATERIALS AND

METHODS

This study was conducted using the clinical data of 408 patients with cardiovascular disease diagnosed with angina pectoris. A coronary computed tomography angiography examination was performed, and quantitative plaque characteristics and pericoronary FAI were analyzed.

RESULTS

Of the 408 patients with angina, 130 had nonobstructive lesions and 278 had obstructive ones. No significant difference in pericoronary FAI was observed between patients with nonobstructive and obstructive lesions. In patients with nonobstructive lesions, the plaque length and pericoronary FAI were significantly higher in patients with unstable angina than in those with stable angina. In patients with obstructive lesions, the plaque fibrolipid volume and percentage were significantly higher in patients with unstable angina than in those with stable angina, and the narrowest lumen area was significantly smaller. Left anterior descending peripheral (peri-LAD) FAI > -83 HU or total plaque length >20.17 mm were independent predictors of unstable angina in patients with nonobstructive lesions. In patients with obstructive lesions, peri-LAD FAI > -77 HU, total lipid volume >12.6 mm3, and narrowest lumen area ≤2.25 mm2 were independent predictors of unstable angina.

CONCLUSION

Pericoronary FAI and total plaque length may be suitable imaging biomarkers for AI-based prediction of the occurrence of unstable angina in patients with nonobstructive lesions.