Slice-based and time-specific hemodynamic measurements discriminate carotid artery vulnerable atherosclerotic plaques
Rui Shen,Xinyu Tong,Dongye Li,Zihan Ning,Hualu Han,Yongjun Han,Dandan Yang,Chenlin Du,Tao Wang,Jingli Cao,Yilan Xu,Ran Huo,Huiyu Qiao,Xihai Zhao,Rui Shen,Xinyu Tong,Dongye Li,Zihan Ning,Hualu Han,Yongjun Han,Dandan Yang,Chenlin Du,Tao Wang,Jingli Cao,Yilan Xu,Ran Huo,Huiyu Qiao,Xihai Zhao
DOI: https://doi.org/10.1016/j.cmpb.2022.107050
IF: 6.1
2022-10-01
Computer Methods and Programs in Biomedicine
Abstract:Background and objectiveHemodynamic patterns play key roles in progression of carotid vulnerable plaques. However, most of previous studies utilized maximum or averaged value of hemodynamic measurements which is not an ideal representative of hemodynamic patterns. This study aimed to investigate the association of slice-based and time-specific hemodynamic measurements with carotid vulnerable plaque using magnetic resonance (MR) vessel wall imaging and histology.MethodsThirty-two patients (mean age: 63.9±8.1 years; 25 males) with carotid atherosclerotic stenosis (≥50% stenosis) referred to carotid endarterectomy were recruited and underwent MR vessel wall imaging. Carotid plaque burden was evaluated on MR images and vulnerable plaque features including calcification, lipid-rich necrotic core, and intra-plaque hemorrhage (IPH) were identified by histology. The slice-based and time-specific hemodynamic measurements were extracted from computational fluid dynamics simulation of 3D carotid arterial model. Correlation coefficients between hemodynamic measurements and carotid plaque features were calculated and the logistic regressions with generalized estimating equation (GEE) were conducted. The value in discriminating carotid vulnerable plaque features was determined by receiver-operating-characteristic analysis.ResultsOf 102 MR-histology matched slices from 32 patients, time-averaged wall shear stress (TAWSS) (r=0.263, p=0.008), oscillatory shear index (OSI) (r=-0.374, p<0.001), and peakWSS (r=0.232, p=0.019) were significantly associated with carotid IPH. The logistic regression with GEE revealed that peakWSS (OR, 1.206; 95% CI, 1.026-1.418; p, 0.023) and TAWSS (OR, 0.364, 95% CI, 0.138-0.959; p, 0.041) were significantly associated with presence of IPH after adjusting for age and BMI. In discriminating carotid IPH, the AUC of TAWSS, OSI, combined TAWSS with maximum wall thickness (MWT) and combined OSI with MWT was 0.656, 0.722, 0.761, and 0.764, respectively.ConclusionsSlice-based and time-specific hemodynamic characteristics could effectively discriminate carotid IPH. Combination of hemodynamic measurements with carotid plaque burden might be a stronger indicator for carotid vulnerable plaque features than each measurement alone.
engineering, biomedical,computer science, interdisciplinary applications,medical informatics, theory & methods