Estimating current and long-term risks of coronary artery in silico by fractional flow reserve, wall shear stress and low-density lipoprotein filtration rate

Background. Hemodynamic changes and consequent low-density lipoprotein (LDL) filtration play an important role in the atherosclerotic plaque development of coronary arteries. In this pilot controlled case study, we aimed to investigate the correlation between parameters derived from computational fluid dynamics (CFD) simulation and risks (both current and long-term) of coronary atherosclerosis. Methods. We reconstructed geometric models from the baseline computed tomography (CT) angiography of two subjects, one patient and one healthy control, and performed CFD simulations. We estimated the current risk of ischemia by fractional flow reserve (FFR). We estimated the potential risk of plaque development by wall shear stress (WSS) and LDL filtration rate with follow-up clinical imaging validation. We investigated the effects of simulation methods (transient/static) and rheological models (Newtonian/Carreau-Yasuda) by comparing the corresponding results (FFR, WSS and LDL filtration rate) in the patient's left anterior descending coronary artery. Results. In baseline CFD simulation, FFR indicated mild current ischemic risk of the patient, in accordance with existing angina pectoris. Baseline WSS and LDL filtration rate results were related with in vivo plaque development. The plaque-growth locations in follow-up CT angiogram coincided with areas of low WSS and high LDL filtration rate in the baseline simulation. The LDL filtration rate delineated more specific risky areas than WSS. Between transient and static results, the difference of FFR was less than 5% in the whole model. As to WSS and LDL filtration rate the transient/static difference was within 20% in most areas, but rose up to 50% for WSS and even higher for LDL filtration rate, in areas with low WSS and high LDL filtration rate. As to rheological effects, Newtonian/Carreau-Yasuda difference was negligible for FFR throughout the model, within 30% for WSS and LDL filtration rate in major areas, and 50% or higher in certain segments where low WSS and high LDL filtration rate existed. Conclusion. CFD results appeared to be related with in vivo development of coronary atherosclerosis. Simulated FFR and its threshold value 0.8 demonstrated the ischemic risk. Both WSS and LDL filtration rate could indicate areas of plaque growth.