Last modified: 2017-05-14
Abstract
Background: Coronary computed tomography angiography-derived fractional flow reserve (FFRCT), which is a noninvasive functional parameter for the diagnosis of coronary artery disease, can solve the clinical problems of invasive fractional flow reserve (FFR) effectively. However, the accuracy is the main problems of FFRCT. The boundary condition is the main factor affecting the accuracy of the FFRCT, and there are few researches about this.
Object: This study explores the personalized settings of the boundary condition in FFRCT in order to improve calculation accuracy.
Methods: In this study, a patient-specific flow boundary condition was presented. Firstly, a vessel volume-based method to calculate flow division fraction was presented over the left anterior descending (LAD) artery, the left circumflex (LCX) artery, and the right coronary artery (RCA) based on the “form-function” relationships. Then, integrated the shear stress formula of the Hagen-Poiseuille flow, the uniform shear hypothesis and Murray’s law, the flow division fraction of each coronary outlet could be calculated. Next, a mathematical model of coronary blood flow at hyperemia () was presented. Some independent physiological parameters of coronary blood flow were selected, including the myocardial mass (), diastolic blood pressure () and heart rate (), the model expressed as
.
Finally, integrated the model and the flow division fraction, the flow rate of each coronary outlet has been calculated. Sixteen cases of patients with coronary stenosis were employed for finite element analyses.
Results: (1) The coronary flow division over LAD, LCX, and RCA was 32.9%, 20.6%, and 46.5% respectively, and they were almost identical to those by the clinical measurement. (2) The mean values of the ratio of total coronary blood flow in cardiac output and myocardial blood flow of 16 patients were 16.97% and 4.07 mL/min/g respectively, in accordance with the rule of medical statistics. (3) The diagnostic accuracy of FFRCT was higher than CT alone (85% vs. 60%) with the reference of clinical FFR, and there was a good agreement between FFRCT and FFR.
Conclusions: The coronary FFRCT has good consistency with invasive FFR under the patient-specific flow boundary condition. This study offers a new way for improving FFRCT accuracy, as well as promotes the clinical application of FFRCT.