A Study of Coronary Bifurcation Shape in a Normal Population

Area of Science:

• Cardiovascular Imaging and Intervention
• Biomedical Engineering
• Computational Anatomy

Background:

• Percutaneous coronary intervention (PCI) uses stents to treat narrowed arteries, but restenosis and stent thrombosis persist.
• Current stent designs may not optimally fit diverse patient anatomies, potentially impacting outcomes.
• Statistical shape analysis offers a novel approach to understand and model coronary artery variations.

Purpose of the Study:

• To develop a statistical shape model of normal coronary artery bifurcations using computed tomographic (CT) angiography data.
• To identify key anatomical features and variations in coronary bifurcations relevant to stent design and deployment.
• To create a computational atlas for improving stent design and predicting clinical outcomes.

Main Methods:

• Utilized CT coronary angiography scans from 211 patients to build statistical shape models of 446 coronary artery bifurcations.
• Employed coherent point drift for registration and principal component analysis (PCA) to quantify shape variations.
• Correlated PCA shape scores with clinical risk factors and compared anatomical laws.

Main Results:

• Identified significant differences in vessel size, bifurcation angles, and curvature across different coronary bifurcations (e.g., left main vs. right coronary crux).
• Observed significant variations in size and bifurcation angle between sexes.
• Found no association between shape and common clinical risk factors like hypertension, smoking, or diabetes. The Huo-Kassab model best fit physiological diameter laws.

Conclusions:

• Coronary bifurcation anatomy exhibits clinically meaningful modes of variation.
• A computational atlas of normal coronary bifurcation shapes can inform the design of next-generation stents.
• This data can enhance bench-top testing and computational fluid dynamics/mechanics modeling for improved PCI strategies.