Evaluation of cerebrovascular hemodynamics in vascular dementia patients with a new individual computational fluid dynamics algorithm
Jian Xie,Zaiheng Cheng,Lihua Gu,Bokai Wu,Gaojia Zhang,Wenshin Shiu,Rongliang Chen,Zan Wang,Chang Liu,Jie Tu,Xiaochuan Cai,Jia Liu,Zhijun Zhang
DOI: https://doi.org/10.1016/j.cmpb.2021.106497
IF: 6.1
2022-01-01
Computer Methods and Programs in Biomedicine
Abstract:BACKGROUND: Cerebral hemodynamic disorders are involved in the occurrence and progression of vascular dementia (VaD), but the methods to detect hemodynamics remainmultifarious and uncertain nowadays. We aim to exploit a computational fluid dynamics (CFD) approach by static and dynamic parameters, which can be used to detect individual cerebrovascular hemodynamics quantitatively.METHODS: A patient-specific CFD model was constructed with geometrical arteries on the magnetic resonance angiography (MRA) and hemodynamic parameters on ultrasound Doppler, by which, the structural and simulated hemodynamic indexes could be obtained, mainly including the cerebral arterial volume (CAV), the number of visible arterial outlets, the total cerebral blood flow (tCBF) index and the total cerebrovascular resistance (tCVR) index. The hemodynamics were detected in subcortical vascular dementia (SVaD) patients (n = 38) and cognitive normal controls (CNCs; n = 40).RESULTS: Compared with CNCs, the SVaD patients had reduced outlets, CAV and tCBF index (all P ≤ 0.001), increased volume of white matter hyperintensity (WMH) and tCVR index (both P ≤ 0.01). The fewer outlets (OR = 0.77), higher Hachinski ischemic score (HIS) (OR = 3.65), increased tCVR index (OR = 1.98) and volume of WMH (OR = 1.12) were independently associated with SVaD. All hemodynamic parameters could differentiate the SVaD patinets and CNCs, especially the composite index calculated by outlets, tCVR index and HIS (AUC = 0.943). Fewer outlets and more WMH increased the odds of SVaD, which were partly mediated by the tCBF index (14.4% and 13.0%, respectively).CONCLUSION: The reduced outlets, higher HIS and tCVR index may be independent risk factors for the SVaD, and a combination of these indexes can differentiate SVaD patients and CNCs reliably. The tCBF index potentially mediates the relationships between hemodynamic indexes and SVaD. Although all simulated indexes only represented the true hemodynamics indirectly, this CFD model can provide patient-specific hemodynamic alterations non-invasively and conveniently.
engineering, biomedical,computer science, interdisciplinary applications,medical informatics, theory & methods