A zero-dimensional predictive model for the pressure drop in the stenotic coronary artery based on its geometric characteristics

• Published in: Journal of biomechanics Vol. 113; p. 110076
• Main Authors: Kim, Jaerim, Jin, Dohyun, Choi, Haecheon, Kweon, Jihoon, Yang, Dong Hyun, Kim, Young-Hak
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 02.12.2020; Elsevier Limited
• Subjects: Angiography; Area; Area reduction and expansion; Blood Flow Velocity; Computational fluid dynamics; Constriction, Pathologic; Convergence; Coronary Angiography; Coronary artery; Coronary Stenosis - diagnostic imaging; Coronary vessels; Coronary Vessels - diagnostic imaging; Curvature; Diameters; Flow separation; Flow velocity; Fluid flow; Geometry; Humans; Learning algorithms; Loss estimation; Machine learning; Mathematical models; Medical imaging; Models, Cardiovascular; Models, Statistical; Navier-Stokes equations; Parameter identification; Partial differential equations; Pipes; Prediction models; Pressure; Pressure drop; Reduction; Reynolds number; Simulation; Stenosis; Stents; Thermal expansion; Three dimensional models; Veins & arteries; Stenosis; Loss estimation; Area reduction and expansion; Pressure drop; Curvature
• ISSN: 0021-9290; 1873-2380
• DOI: 10.1016/j.jbiomech.2020.110076

The diameter- or area-reduction ratio measured from coronary angiography, commonly used in clinical practice, is not accurate enough to represent the functional significance of the stenosis, i.e., the pressure drop across the stenosis. We propose a new zero-dimensional model for the pressure drop across the stenosis considering its geometric characteristics and flow rate. To identify the geometric parameters affecting the pressure drop, we perform three-dimensional numerical simulations for thirty-three patient-specific coronary stenoses. From these numerical simulations, we show that the pressure drop is mostly determined by the curvature as well as the area-reduction ratio of the stenosis before the minimal luminal area (MLA), but heavily depends on the area-expansion ratio after the MLA due to flow separation. Based on this result, we divide the stenosis into the converging and diverging parts in the present zero-dimensional model. The converging part is segmented into a series of straight and curved pipes with curvatures, and the loss of each pipe is estimated by an empirical relation between the total pressure drop, flow rate, and pipe geometric parameters (length, diameter, and curvature). The loss in the diverging part is predicted by a relation among the total pressure drop, Reynolds number, and area expansion ratio with the coefficients determined by a machine learning method. The pressure drops across the stenoses predicted by the present zero-dimensional model agree very well with those obtained from three-dimensional numerical simulations.