The importance of assessing the collateral status (CS) in patients with ischaemic stroke (IS) has repeatedly been emphasised in clinical guidelines. Various publications offer qualitative or semiquantitative scales with gradations corresponding to the different extents of the collaterals, visualised mostly on the basis of CTA images. However, information on their inter-rater reliability is limited. Therefore, the aim of this study is to investigate the inter-rater reliability of the scales for collateral assessment. CTA images of 158 patients in the acute period of IS were used in the study. The assessment of CS was performed by two experts using three methodologies: the modified Tan scale, the Miteff scale, and the Rosenthal scale. Cohen’s kappa, weighted kappa and Krippendorff’s alpha were used as reliability measures. For the modified Tan scale and the Miteff and Rosenthal scales, the weighted kappa values were 0.72, 0.49 and 0.59, respectively. Although the best measure of consistency was found for the modified Tan scale, no statistically significant differences were revealed among the scales. The impact of the CS on the degree of neurological deficit at discharge was shown for the modified Tan and Rosenthal scales. In conclusion, the analysis showed a moderate inter-rater reliability of the three scales, but was not able to distinguish the best one among them.

Оценка фракционного резерва кровотока по данным компьютерной томографии: сравнение расчетных показателей с результатами инвазивных измерений

The authors performed clinical studies based on modelling of an ascending aortic aneurysm in 37 patients and 10 apparently healthy subjects. Echocardiography was carried out in the B-mode using the Vivid E9 device (USA, GE). The linear dimensions of the aorta were assessed at three points - in the immediate vicinity of the valves, in the area of the maximum dilatation and in the area of decreased dilatation with registration of blood flow velocity in the aorta. The aortic walls were contoured with the division of equal intervals into 4 portions in order to obtain longitudinal shear deformation velocity during the cardiac cycle. We worked out a system of assessing the velocity vector fields with the help of transthoracic echocardiography in patients with an ascending aortic aneurysm, based on registration of blood flows, which made it possible to obtain the components of velocity. We also determined an optimal method of assessing turbulence in the aorta taking into account the direction of the vectors. Obtained were the numerical data of aortic wall deformation velocity in the longitudinal direction and calculation of the weighting function with the distinction between pathology and the norm. Based on the deformation, the distance between the registered points, and the movement of the vascular wall, we determined the reference values of blood flow velocity inside the aorta and immediately close to its walls.

Assessment of rupture risk for intracranial aneurysms (IA) is a particular challenge in cases of so-called complex aneurysms due to their variable morphometric characteristics. Arterial branch arising from the dome or the neck of IA is one of the least explored features of complex aneurysms. The methods of computational fluid dynamics may be valuable to determine the influence of arterial branches of IA on local hemodynamics. To analyze local hemodynamics in IA with arterial branch arising from the cupola or the neck depending on the structure of the aneurysm and blood flow rate in the parent vessel. CT angiography data of 4 patients with IA were estimated in this study. Modifications of the baseline 3D models of the aneurysms resulted 12 patient-specific models included into analysis. Hemodynamic calculations were made by using of ANSYS Workbench 19 software package. Wall shear stress (WSS) was characterized by the most significant variability, especially in case of sidewall aneurysms. Small cross-sectional area of additional branch in relation to the neck of IA was not followed by considerable changes of blood flow patterns inside IA after «virtual» removal of the vessel. Otherwise, the intensity of flows was drastically reduced. Simulation of high inlet flows demonstrated substantial variation of WSS in the area of jet. Additional arterial branch arising from the dome or the neck of IA significantly influences local hemodynamics. This influence depends on the localization of IA in relation to the parent vessel and the diameter of additional arterial branch.

Intracranial aneurysms (IAs) pose a high risk of spontaneous subarachnoid hemorrhage. In the most complex cases, the only way to exclude the aneurysm from the circulation is to perform a high-flow extracranial-to-intracranial bypass, thus creating a new bloodstream. This avoids severe ischemic complications; however, it requires careful consideration of individual anatomy and hemodynamic parameters. Computational fluid dynamics (CFD) can be of great help in planning such a surgery by creating 3D patient-specific models of cerebral circulation. Assessment of the perspectivity of high-flow extracranial-to-intracranial bypass planning using computational modeling. In this research work, we have applied the CFD methods to a patient with a giant thrombosed IA of the internal carotid artery (ICA). Preoperative CTA images and Gamma Multivox workstation were used to create a 3D model with current geometry and three additional models: Normal anatomy (no IA), Occlusion (with ligated ICA), Virtual bypass (with bypass and ligated ICA). The postoperative data were also available. Boundary conditions were based on PC-MRI measurements. Calculation of hemodynamics was conducted with a finite element package ANSYS Workbench 19. The results demonstrated an increase in the blood flow on the affected side by more than 70% after the virtual surgery and uniformity of flow distribution between the affected and contralateral sides, indicating that the treatment is likely to be efficient. Later, postoperative data confirmed that. The study showed that virtual preoperative CFD modeling could significantly simplify and improve surgical planning.

In ischemic stroke, the condition of cerebral collateral circulation is one of the key factors determining the outcome. Digital subtraction angiography is considered the gold standard of evaluation of cerebral collateral circulation. However, computed tomography angiography is the most widely used method characterized by high level of conformity with subtraction angiography. Currently, several scales of visual evaluation of collateral circulation development in images obtained by computed tomography angiography are used. The scales describe the territory of stroke-associated cerebral artery, as well as details of various areas of the territory. The association between the score and severity of neurological deficit and volume of cerebral ischemia was demonstrated in numerous studies. However, consensus on the most reliable method of description of cerebral collateral status has not been reached. Use of modern methods of processing of medical images and artificial intelligence allowed to make a significant step towards automatization of collateral status evaluation with such benefits as high processing speed and resistance to subjective opinion. Despite low number of studies on this subject, implementation of automated solutions has already showed its effectiveness. In this review, scales for manual evaluation of cerebral collateral status are considered, their reliability is described, and current approaches to automated evaluation of collateral circulation in ischemic stroke are presented.