INTRODUCTION
Vessel tortuosity is a frequently observed angiographic feature, manifesting across various organ systems [3,12]. Vessel tortuosity affects blood flow, whereas hemodynamic stress induced by intraluminal flow into the vessel walls influences tortuosity [6]. In this interaction, the blood vessels deform in the direction of the intraluminal flow to reduce hemodynamic stress [6]. In systemic diseases, such as arterial hypertension and diabetes mellitus, vessel tortuosity is aggravated [3,12]. Notably, an increase in vessel tortuosity causes a change in hemodynamic stress, which can cause vessel wall impairment, resulting in the formation or rupture of cerebral aneurysms [6,13,15].
Several studies on the association between vessel tortuosity and intracranial aneurysms have been frequently published [1,4-11,16,18,19]. However, parameters and standard imaging modalities that represent vessel tortuosity have not yet been established. In most studies, vessel tortuosity investigators rely on the analysis of digital subtraction angiography (DSA) and magnetic resonance angiography (MRA) images [1,4-11,19]. However, neither modality has been universally endorsed as a standard imaging tool for such assessments. In this study, we compared the tortuosity through the vessel angle using MRA and DSA measurements in the same patients.
MATERIALS AND METHODS
This was a retrospective study. The study protocol was approved by our institutional review board, and the requirement for written informed consent from included patients was waived. A total of 85 patients were enrolled between December 2021 and December 2022. All patients who underwent DSA for the evaluation of unruptured aneurysms in the anterior circulation and those who also underwent MRA were included in this study.
MR imaging protocol
Three-dimensional (3D) unenhanced time-of-flight (TOF) MRA images were acquired on a 3T scanner (Ingenia Elition X 3T; Philips Medical System, Best, Netherlands) with the following parameters: field of view 220×220 mm, scan size 640×324, 180 1.2 mm slices, repetition time (TR) of 20 ms, echo time (TE) of 3.45 ms, flip angle (FA) of 20°, reduction factor 4.5, acquired voxel size 0.344×0.68×1.2 mm3, reconstructed voxel size 0.344×0.344×0.6 mm3.
DSA imaging protocol
Three-dimensional rotational angiography (3DRA) images were obtained using biplane C-arm digital angiography equipment (Allura Clarity FD 20/20, Philips Medical Systems, Netherlands) with a 20-inch field of view, frame rate of 30 f/s, and rotation range of 270°.
The contrast medium (Visipaque-320; GE Healthcare, Cork, Ireland) was injected at a flow rate of 4 mL/s for a total of 7 mL via a 5F catheter positioned at the cervical portion of the internal carotid artery for DSA imaging. For 3D DSA imaging, the contrast medium was injected at a flow rate of 3 mL/s for a total of 18 mL, followed by a scan delay of 2 s.
RA was performed with a 270° rotation of the C-arm in 4.1 seconds. The imaging data were automatically exported to a workstation. For rotational angiography, the pressure of the contrast medium injected into the artery through the catheter was adjusted to ensure that the maximum value did not exceed 500 dpi.
Measurement of vessel angles
The measurements of vessel angles were performed by a researcher who was blinded to the clinical information and study design. Fig. 1 illustrates the study process. To compare the differences in angles between MRA and DSA in three dimensions, MRA and DSA data for each patient were reconstructed as 3D models using digital imaging and communications in medicine (DICOM) images. Subsequently, the center points of the models were calculated using a vascular modeling toolkit (VMTK) devised by Izzo et al. [17] to minimize bias in manual measurement. Finally, the MRA and DSA morphologies were compared by measuring three angles in the internal carotid artery (ICA) bifurcation segment and two angles in the carotid siphon segment. These angles included the angles between the ICA and middle cerebral artery (MCA) (ICA - MCA), between the ICA and anterior cerebral artery (ACA) (ICA - ACA), MCA and ACA (MCA - ACA), the proximal carotid siphon angle, and distal carotid siphon angle. The angles in ICA - Bifurcation segment were measured on the intersected longitudinal centerlines of the two vessels of interest (Fig. 2A) [18]. Additionally, the proximal and distal carotid siphon angles were measured at the intersection of two lines traced through the center points in each straight segment of the siphon (Fig. 2B) [17]. The commercial computer-aided design software (CATIA, V5 - 6R2012, Dassault Systems; Meshmixer, version 11.0.544, Autodesk) was used to calculate the angles and reconstruct the 3D model. Furthermore, the inflow angle was measured at the intersection point between the centerline of the parent vessel and the line from the center of the neck to the tip of the aneurysm dome (Fig. 2C) [1].
Statistical analysis
Categorical data are presented as frequencies and percentages (%), and continuous data are expressed as means and standard deviations. The normality of the variables was assessed using the Shapiro-Wilk test. The variables that demonstrated normality include ICA-MCA, distal and inflow angle measured by DSA, and flow angle measured by MRA. Due to the correlation of data collected from each patient, the differences among the variables were analyzed. Variables that met normality were presented as means and standard deviations, and paired t-tests were used for parametric methods. Variables that did not meet normality are presented as median and interquartile range (IQR), and the Wilcoxon signed-rank test was used for nonparametric methods. Pearson’s correlation analysis was performed to determine the correlation between variables, and the correlation coefficients and p-values are shown. The statistical analysis was performed using the SAS program (version 9.4; SAS Institute, Cary, North Carolina, USA); unless otherwise stated, all tests were performed at a significance level of 0.05.
RESULTS
This study included 85 patients, of whom 64 (75.3%) were males. The mean (±standard deviation) age of the study group was 61.5±13.5 years, ranging from 26 to 87 years. A total of 47 patients (55.3%) had hypertension, 12 (14.1%) had diabetes, 49 (57.7%) had hyperlipidemia, 16 (18.8%) had a history of alcohol consumption, and 11 (12.9%) had a history of smoking. In addition, a history of stroke was identified in seven patients (8.2%).
Table 1 compares the vessel angles measured using MRA and DSA at the same vessel location in the same patients. No difference in the vessel angles was observed between the measurement methods at any location, except for the ICA - MCA location, where a statistically significant difference was observed. Specifically, the median value of the ICA - MCA angle was 50.75˚ (IQR, 34.91-62.24) when measured by DSA and 50.26˚ (IQR, 33.49-70.57) when measured by MRA. The p-value was 0.035.
The inflow angles measured by MRA and DSA were not significantly different (Table 1).
DISCUSSION
In recent literature, several studies have highlighted the impact of vessel tortuosity and inflow angle on the formation and rupture of intracranial aneurysms (Table 2) [1,4-11,16,18,19]. These previous studies used various parameters to establish a connection between increased tortuosity and the presence of intracranial aneurysms. A similar association was found for the ICA, MCA, anterior communicating artery (Acom), and vertebral artery (VA). However, the modalities used for the cerebral artery measurements varied among different studies. Several authors used DSA, whereas others used MRA to measure the cerebral arteries. In this study, the significant difference in ICA - MCA angle between MRA and DSA was found while no discrepancies were found in other vessel angles. In addition, it was confirmed that there were no differences in the parameters including inflow angles between MRA and DSA.
DSA is the most accurate modality for measuring the shape or size of aneurysms and is considered the gold standard for aneurysm diagnosis. However, the vessel tortuosity of the actual cerebral artery, indicated by the vessel angle, may differ when measured using DSA. This discrepancy could arise because DSA images of cerebral artery were obtained by injecting contrast medium directly into the artery with artificial pressure, potentially causing a slight change in vessel angle. Conversely, because computed tomography angiography (CTA) is performed while injecting contrast agents into veins, no artificial pressure is applied directly to the cerebral arteries. In addition, MRA can obtain the shape of the cerebral arteries without using contrast agents using TOF MRA techniques. Therefore, CTA and MRA are considered more physiologically natural than DSA.
The results of this study suggest the possibility of discrepancies in vessel angles depending on the measurement modality, particularly in certain regions of cerebral arteries. Comparing the DSA using X-ray, the TOF-MRA is sensitive to flow-related artifacts, signal loss related to saturation effect at the margin of the slab, and atherosclerotic changes in the vessel [2]. Therefore, discerning the modality used when measuring vessel tortuosity and inflow angle is crucial.
The results of this study show that the discrepancy in the vessel angle in the same patient was significant for the ICA-MCA angle. This can be explained by the distinctive angioarchitecture of the MCA, which is less fixed because it is far from the skull. In other words, the ICA is attached to the skull by the dural ring and cavernous sinus in the paraclinoid segment, whereas distal arteries, such as the ACA and MCA, are less fixed in the subarachnoid space. In this study, the reason for the lack of difference in the ICA - ACA vessel angle compared with the difference in the ICA - MCA vessel angle according to the measurement methods is presumed to be that the ACA is connected to the contralateral ACA by the anterior communicating artery.
A difference in the accuracy was also observed depending on the use of two-dimensional (2D) images or 3D images. Several authors have devised various parameters to analyze the vessel tortuosity of various segments of the cerebral artery and have investigated the parameters associated with the formation of intracranial aneurysms. They used 2D DSA images for measurement. However, because cerebral arteries are three-dimensionally distributed the angle of the blood vessels must also be measured in 3D space. Skodvin et al. argued that in 2D images, angles depended on the viewing plane from which the observer saw the aneurysm [14]. For this reason, vessel angles were measured between vectors in 3D space to ensure precise and reproducible measurements independent of the viewing plane [14]. In this study, the 3D vessel was reconstructed with the DICOM files of DSA and MRA to measure the angle in the 3D space, so the inaccuracy of the measurement method in 2D could be overcome.
This study had several limitations. First, the retrospective nature of the study lends itself to gaps in prospective data collection. Secondly, the sample size of each cerebral artery segment was limited, which did not allow for a more powerful analysis. Further studies with larger sample sizes are needed to confirm our findings. Finally, a possibility of an inherent limitation exists in the measurement method for the vessel angle. Even if we applied the measurement methods used in previous studies, we cannot eliminate the concern that the method may not accurately measure the vessel angle. Additionally, patient comorbidities such as hypertension and diabetes, which can affect vessel tortuosity, must also be considered during hemodynamic studies, and future research on this will need to be expanded.
CONCLUSIONS
Given the characteristics of diagnostic modalities, DSA is considered the most accurate method for depicting the precise morphology of cerebral arteries and aneurysms. Our study found a discrepancy between MRA and DSA in measuring the angle between the ICA and the MCA. Since vessel tortuosity significantly affects the results of hemodynamic simulations, it’s essential to take our findings into account when analyzing cerebral vessels using MRA. And, further research is needed to understand the reasons for the discrepancies between MRA and DSA, as well as to investigate technologies that could help minimize these differences. These efforts will greatly improve our understanding of cerebral vascular structures and vessel tortuosity, ultimately enhancing hemodynamic studies.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
