Diagnostic performance of a vessel-length-based method to compute the instantaneous wave-free ratio in coronary arteries.

The instantaneous wave-free ratio (iFR) is a recently introduced vasodilator-free index to assess the functional severity of coronary stenosis in the resting state, while fractional flow reserve (FFR) is the gold standard index in hyperemia. The computed instantaneous wave-free ratio (CT-iFR) is a noninvasive method to estimate iFR using computer simulations. Here, we developed a vessel-length-based CT-iFR method in patient-specific models of coronary arteries. This method was implemented by coupling a three-dimensional computational fluid dynamics model with a lumped parameter model (LPM) of coronary circulation in a non-hyperemic resting state. A time-varying resistance in the LPM was used for the iFR simulation. In total, 50 coronary vessels of 32 patients were computed, and their CT-iFR values were compared with clinically measured iFRs to evaluate the diagnostic performance of the present CT-iFR method. The area under the receiver operating characteristics curve of CT-iFR validation was 0.93. In diagnostic performances of CT-iFR, accuracy, sensitivity, and specificity were 86%, 83.3%, and 86.8%, respectively. These results indicate that this CT-iFR method can be used as a pre-operative aid to establish a percutaneous coronary intervention strategy as a noninvasive alternative to iFR.

Prediction of Plaque Progression in Coronary Arteries Based on a Novel Hemodynamic Index Calculated From Virtual Stenosis Method.

RATIONALE: Predicting the sites in coronary arteries that are susceptible to plaque deposition is essential for the development of clinical treatment strategies and prevention. However, to date, no physiological biomarkers for this purpose have been developed. We hypothesized that the possibility of plaque deposition at a specific site in the coronary artery is associated with wall shear stress (WSS) and fractional flow reserve (FFR). BACKGROUND AND OBJECTIVE: We proposed a new biomarker called the stenosis susceptibility index (SSI) using the FFR and WSS derived using virtual stenosis method. To validate the clinical efficacy of this index, we applied the method to actual pilot clinical cases. This index non-invasively quantifies the vasodilation effects of vascular endothelial cells relative to FFR variation at a specific coronary artery site. METHODS AND RESULTS: Using virtual stenosis method, we computed maximum WSS and FFR according to the variation in stenotic severity at each potential stenotic site and then plotted the variations of maximum WSS (y-axis) and FFR (x-axis). The slope of the graph indicated a site-specific SSI value. Then we determined the most susceptible sites for plaque deposition by comparing SSI values between the potential sites. Applying this method to seven patients revealed 71.4% in per-patient basis analysis 77.8% accuracy in per-vessel basis analysis in percutaneous coronary intervention (PCI) site prediction. CONCLUSION: The SSI index can be used as a predictive biomarker to identify plaque deposition sites. Patients with relatively smaller SSI values also had a higher tendency for myocardial infarction. In conclusion, sites susceptible to plaque deposition can be identified using the SSI index.

A patient-specific virtual stenotic model of the coronary artery to analyze the relationship between fractional flow reserve and wall shear stress.

OBJECTIVE: As the stenotic severity of a patient increases, fractional flow reserve (FFR) decreases, whereas the maximum wall shear stress (WSSmax) increases. However, the way in which these values can change according to stenotic severity has not previously been investigated. The aim of this study is to devise a virtual stenosis model to investigate variations in the coronary hemodynamic parameters of patients according to stenotic severity. METHODS: To simulate coronary hemodynamics, a three-dimensional (3D) coronary artery model of computational fluid dynamics is coupled with a lumped parameter model of the coronary micro-vasculature and venous system. RESULTS: To validate the present method, we first simulated 13 patient-specific models of the coronary arteries and compared the results with those obtained clinically. Then, virtually narrowed coronary arterial models derived from the patient-specific cases were simulated to obtain the WSSmax and FFR values. The variations in FFR and WSSmax against the percentage of diameter stenosis in clinical cases were reproducible by the virtual stenosis models. We also found that the simulated FFR values were linearly correlated with the WSSmax values, but the linear slope varied by patient. CONCLUSION: We implemented 130 additional virtual models of stenosed coronary arteries based on data from 13 patients and obtained statistically meaningful results that were identical to the large-scale clinical studies. And the slope of the correlation line between FFR and WSSmax may help clinicians to design treatment plans for patients.

A novel patient-specific model to compute coronary fractional flow reserve.

The fractional flow reserve (FFR) is a widely used clinical index to evaluate the functional severity of coronary stenosis. A computer simulation method based on patients' computed tomography (CT) data is a plausible non-invasive approach for computing the FFR. This method can provide a detailed solution for the stenosed coronary hemodynamics by coupling computational fluid dynamics (CFD) with the lumped parameter model (LPM) of the cardiovascular system. In this work, we have implemented a simple computational method to compute the FFR. As this method uses only coronary arteries for the CFD model and includes only the LPM of the coronary vascular system, it provides simpler boundary conditions for the coronary geometry and is computationally more efficient than existing approaches. To test the efficacy of this method, we simulated a three-dimensional straight vessel using CFD coupled with the LPM. The computed results were compared with those of the LPM. To validate this method in terms of clinically realistic geometry, a patient-specific model of stenosed coronary arteries was constructed from CT images, and the computed FFR was compared with clinically measured results. We evaluated the effect of a model aorta on the computed FFR and compared this with a model without the aorta. Computationally, the model without the aorta was more efficient than that with the aorta, reducing the CPU time required for computing a cardiac cycle to 43.4%.