Abstract
ELAPSS score is commonly utilized for predicting the growth of unruptured cerebral aneurysms. However, its application is unsuitable for small aneurysms with high demand for growth prediction. Consequently, we investigated the diagnostic accuracy of semi-quantitative assessment using the ELAPSS score and hemodynamic parameters using computational fluid dynamics in small aneurysms. A prospective observational study from January 2013 to February 2022 included 185 patients with 215 unruptured aneurysms with a maximum diameter of 3-5 mm. Aneurysms were classified into stable (186 aneurysms) and growth (29 aneurysms) groups based on repeated images. We calculated high shear area ratios, high shear concentration ratios, and flow concentration ratios as hemodynamic parameters that we have already reported to be associated with small aneurysm growth in our previous study. The characteristics associated with the growth of small aneurysms were statistically investigated with morphological variables and hemodynamic parameters. The ELAPSS score was also calculated for the same aneurysm group to determine whether the growth risk was sufficiently assessed. In morphological variables, no significant differences were observed between the 2 groups. As for the hemodynamic parameters, the growth group had a significantly lower flow concentration ratio (0.61 vs 0.66, p = 0.016), lower high shear area ratio (0.28 vs 0.33, p < 0.001), and a higher high shear concentration ratio (6.39 vs 5.01, p < 0.001). However, there were no significant differences in the ELAPSS scores between the 2 groups. When limited to small aneurysms, computational fluid dynamics may offer more enhanced predictive capabilities compared to the ELAPSS score for identifying growth tendencies.
Keywords: computational fluid dynamics, small unruptured cerebral aneurysms, high shear concentration ratio
Introduction
The growth of unruptured cerebral aneurysms is strongly associated with rupture. Inoue et al.1) reported that the risk of rupture was 18.5% per person-year following aneurysm growth. Chmayssani et al.2) found that 90% of aneurysms smaller than 7 mm enlarged prior to rupture. Consequently, preventing aneurysm rupture becomes plausible if growth prediction can be conducted in advance. Despite of the general use of ELAPSS score to predict aneurysm growth,3) the original study included approximately 40% of patients in the growth group who had aneurysms with a maximum diameter of 5 mm or greater warranting surgical intervention regardless of whether the aneurysm had visibly enlarged or not.3-5) The ELAPSS score is not restricted to small aneurysms with a high demand for growth prediction. Thus, the growth prediction method of aneurysms that is considered most necessary and useful in decision-making should be developed for small aneurysms with a maximum diameter of less than 5 mm. In recent years, several reports have explored aneurysm growth using computational fluid dynamics (CFD),6-8) particularly focusing on small aneurysms with a maximum diameter of 3 to 5 mm.8) In this study, we investigated the diagnostic accuracy of both the ELAPSS score and hemodynamic parameters using CFD for growth prediction limited to small aneurysms.
Materials and Methods
This prospective study was approved by the Ethics Committee at Mie University, Tsu, Mie, Japan (approval number 2469). This study was undertaken to conform to the Systematic Multicenter Study of Unruptured Cerebral Aneurysms based on the Rheological Technique at Mie (SMART Mie) protocol. The joint research facilities were 11 hospitals in Japan. Written informed consent was obtained from all patients prior to the enrollment in the study.
Patient population and study protocol
This study's participants comprised patients enrolled in the SMART registry between January 2013 and February 2022. This study was designed to include individuals meeting the following criteria: 1) those with intracranial unruptured saccular aneurysms ranging in diameter of 3-5 mm confirmed by magnetic resonance angiography (MRA) at the time of registry enrollment; 2) individuals who had undergone computed tomography angiography (CTA), or digital subtraction angiography (DSA) for subsequent CFD; 3) individuals of independently performing activities of daily living (modified Rankin Scale scores 0-2); and 4) cases with repeated MRA imaging evaluation.
Patients were categorized into stable and growth groups based on repeated imaging studies with the initial MRA serving as the reference. The growth of aneurysms was assessed using 3-dimensional (3D) time-of-flight (TOF) MRA. In both the initial and follow-up studies, 2 experienced neurosurgeons independently compared the 3D TOF MRA findings with those derived from CFD in a blinded manner. In the previous report,8) growth was defined as 0.5 mm or greater;8) however, in the present study, growth was defined as 1.0 mm or greater, in concordance with an indisputable change consisting of the evaluation method in the ELAPSS score.3) Aneurysms exhibiting no morphological changes on imaging after at least 1 year of follow-up were categorized as stable. In instances of evaluator disagreement, a joint reevaluation was conducted to reach a consensus.
CFD analysis
The CFD analysis was conducted without knowledge of the aneurysms' growth information. We generated patient-specific models using either the Digital Imaging and Communications in Medicine (DICOM) format with 3D CTA or a virtual reality modeling language (VRML) data set with 3D DSA. Data from DICOM were imported into a 3D modeling software (Mimics Innovation Suite version 24.0, Materialise Japan), from which vascular geometry was exported as a stereolithography (STL) file. Referring to the 3D vessel geometry in VRML and STL, vessels with a diameter of less than 1 mm were excluded from the analysis, and then the cerebral arterial wall was remeshed for the purpose of improving the quality of the surface triangle and smoothening the surface of the vascular geometry using commercial software (3-matic version 16.0, Materialise Japan). The computational hybrid meshes were generated using tetrahedral and prism elements (ICEM CFD 2021 R2, ANSYS Inc.). Tetrahedral elements ranged in size from 0.1 to 0.6 mm. To ensure an accurate definition of the velocity gradient locally, 6 prismatic boundary layers were applied to the vessel wall with a total thickness of 0.15 mm. The 3D incompressible laminar flow field was obtained by solving the continuity and Navier-Stokes equations. Numerical modeling was carried out using a commercially available package for CFD (CFX 2021 R2, ANSYS Inc.). Blood density was assumed to be 1,056 kg/m3 and blood dynamics viscosity to be 0.0035 Pa・s. To achieve physiological wall shear stress (WSS).9,10) The internal carotid artery and vertebral artery flow waveforms were used for transient analysis. Traction-free boundary conditions were applied at the outlets. Utilizing the initial value specification from the preceding steady-state analysis, one pulsatile cycle was taken as output with time steps of 0.0001 s.
Morphological variables
The same technique used in the previous report was applied to the measurement of two-dimensional morphological variables derived from 3D CTAs or 3D DSAs generated using STL with ImageJ software (National Institutes of Health).11) As a 3D morphological variable, the volume-to-ostium area ratio (VOR) was measured using the CFX 2021 R2 (ANSYS Inc.).12)
Hemodynamic parameters
According to the previous investigation, the growth of small aneurysms was associated with a lower inflow jet and a local concentration of higher WSS.8) In other words, the hemodynamic parameters characteristic of small aneurysm growth were lower flow concentration ratio (FCR), lower high shear area (HSA) ratio (HSAR), and higher high shear concentration ratio (HSCR).8) High WSS (HWSS) required for the calculation of HSCR was defined as 110% of the time-averaged WSS over the entire dome. HSA represented regions with values above HWSS, while HSAR denoted the ratio of HSA to the surface area of the dome. The calculation methods of HSCR and FCR were the same as those in previous reports.8) In this study, we redefined aneurysm growth from 0.5 mm to 1.0 mm to align with the ELAPSS score,3,8) and these hemodynamic parameters were reevaluated. Additionally, we investigated flow velocity and WSS-related hemodynamic parameters that had been reported to be associated with aneurysm initiation, growth and rupture.13-19)
ELAPSS score
ELAPSS score was developed based on 10 cohort studies of 1,909 unruptured cerebral aneurysms to estimate the risk of aneurysm growth at 3 and 5 years.3) Aneurysms were scored based on 6 criteria: earlier subarachnoid hemorrhage, location of the aneurysm, age, population, size of the aneurysm, and shape of the aneurysm.3) For the age criterion, one point was added every 5 years over 61 years. As the oldest patient in this study was 86 years old, a maximum of 6 points was added. The only subjective criteria among the 6 was the shape of aneurysms. For the regular shape, no points were added, while for the irregular shape, 4 points were added. In the original paper, the irregular shape was defined as the presence of blebs, aneurysm wall protrusions, or multiple lobes.3) However, the original paper and references did not provide a detailed description of irregular aneurysms.3,20,21) Hence, standardizing the evaluation of aneurysm shape for this study was necessary. At the neck of each aneurysm, 2 horizontal cross-sectional images were created using CFX-post (CFX 2021 R2, ANSYS Inc.). These images were positioned optimally to separate the aneurysm from the parent artery and its branches and rotated 90 degrees along the neck. Based on the 2 planar images, 3 typical images of irregularly shaped aneurysms were defined (Fig. 1). Two experienced neurosurgeons, serving as blinded examiners, evaluated the shapes of the aneurysms presented with images of irregularly shaped aneurysms. In cases of conflicting evaluations, the aneurysm shapes were reevaluated collaboratively by both observers, leading to a consensus.
Fig. 1.
Two orthogonal appearances of a representative aneurysm of irregular shape.
The presence of bleb (A-B), aneurysm wall protrusion (C-D), and multiple lobes (E-F). A, C, and E show optimal planes for the separation of the aneurysm from the parent artery and its branches. B, D, and F show planes after the rotation of the optimal plane by 90 degrees along the neck.
Statistical analysis
All values were presented as a median (interquartile range [IQR]) and statistical analysis was conducted using EZR (Easy R).22) For assessing the statistical significance of differences between the stable and growth groups, the Brunner-Munzel and Pearson chi-square tests were employed. A p < 0.05 was considered statistically significant. The diagnostic accuracy of predicting aneurysm growth was evaluated using the area under the receiver operating characteristic (ROC) curve. The reproducibility of the inter-observer evaluation of the irregular shape of an aneurysm was quantified by the kappa test.
Results
During the study period, the registry enrolled 718 patients with unruptured cerebral aneurysms. In a total of 434 patients, 3D CTA and DSA images required for CFD analyses were lacking, or surgical intervention was performed without repeated imaging studies. Among 338 aneurysms in 284 patients, 117 were excluded due to a maximum diameter exceeding 5 mm. Additionally, 6 aneurysms for which repeated MRA imaging could not be performed were excluded. Ultimately, we analyzed 215 aneurysms in 185 patients who met the inclusion criteria. All patients were Japanese. The median observation period for all aneurysms was 1,099 days (IQR 725-1777).
Univariate analysis
A brief description of each group's background data can be found in Table 1. The results revealed that the growth group had a higher incidence of hypertension than the stable group. In morphological variables, no significant differences were observed between the stable and growth groups (Table 2). Regarding the hemodynamic parameters, the growth group had significantly lower FCR (0.61 vs 0.66, p = 0.016), lower HSAR (0.28 vs 0.33, p < 0.001), and higher HSCR (6.39 vs 5.01, p < 0.001) (Table 2, Fig. 2). However, ELAPSS scores did not differ significantly between the 2 groups (Table 3). In terms of the inter-observer reproducibility in the evaluation of irregular aneurysms, the kappa value was 0.273 (0.111-0.435).
Table 1.
Patient and aneurysm characteristics in the stable and growth groups
| All | Stable group | Growth group | p Value* | |
|---|---|---|---|---|
| No. of patients | 185 | 156 | 29 | |
| Women | 134 | 112 | 22 | Ref |
| Men | 51 | 44 | 7 | 0.822 |
| No. of aneurysms | 215 | 186 | 29 | |
| Follow up period, days | ||||
| Median | 1,099 | 1,101 | 1,086 | |
| Age, years | ||||
| Median | 64 | 64 | 64 | |
| < 61 | 68 | 57 | 11 | Ref |
| ≥ 61 | 117 | 99 | 18 | 0.453 |
| Single aneurysm | 125 | 105 | 18 | Ref |
| Multiple aneurysms† | 60 | 51 | 11 | 0.669 |
| Site of aneurysms | ||||
| ICA | 102 | 87 | 15 | Ref |
| MCA | 59 | 51 | 8 | 1.000 |
| ACA | 35 | 34 | 1 | 0.070 |
| VABA | 19 | 14 | 5 | 0.309 |
| Aneurysm subtype | ||||
| Sidewall type | 96 | 84 | 12 | Ref |
| Bifurcation type | 119 | 102 | 17 | 0.841 |
| Concomitant Disease | ||||
| Hypertension | 93 | 72 | 21 | 0.014 |
| No hypertension | 92 | 84 | 8 | Ref |
| Previous SAH | 13 | 9 | 4 | 0.126 |
| No previous SAH | 172 | 147 | 25 | Ref |
| Family history | ||||
| Family history of SAH | 19 | 18 | 1 | 0.317 |
| No family history of SAH | 166 | 138 | 28 | Ref |
| Social history | ||||
| Current smoker | 21 | 18 | 3 | 1.000 |
| No current smoker | 164 | 138 | 26 | Ref |
Values are presented as number of patients or aneurysms unless otherwise indicated. *The p values are calculated using Pearson’s chi-square tests between the stable and growth groups. †If one patient has two aneurysms, including one aneurysm larger than 5 mm, it is counted as n = 1 for aneurysms in this study. ACA indicates anterior cerebral artery; ICA, internal cerebral artery; MCA, middle cerebral artery; SAH, subarachnoid hemorrhage; and VABA, vertebral artery and basilar artery
Table 2.
Morphological variables and hemodynamic parameters of aneurysms in the stable and growth groups
| Stable group (n=186) | Growing group (n=29) | p Value* | |
|---|---|---|---|
| Morphological variables | |||
| Aneurysm Depth, mm | 2.33 (1.71 - 2.75) | 2.40 (1.92 - 2.95) | 0.268 |
| Projection Length, mm | 2.40 (1.78 - 2.84) | 2.51 (2.24 - 3.07) | 0.136 |
| Maximum Size, mm | 3.89 (3.41 - 4.40) | 4.03 (3.42 - 4.61) | 0.331 |
| Neck Width, mm | 3.45 (3.05 - 3.96) | 3.16 (2.98 - 3.95) | 0.677 |
| Parent Artery Diameter, mm | 2.90 (2.38 - 3.81) | 3.00 (2.50 - 3.28) | 0.932 |
| Aspect Ratio | 0.65 (0.49 - 0.80) | 0.71 (0.56 - 0.86) | 0.118 |
| Projection Ratio | 0.67 (0.51 - 0.82) | 0.74 (0.59 - 0.88) | 0.079 |
| Size Ratio | 1.33 (1.06 - 1.56) | 1.37 (1.18 - 1.61) | 0.471 |
| Dome Area, mm2 | 22.25 (15.52 - 29.93) | 23.52 (17.69 - 31.37) | 0.188 |
| Neck Area, mm2 | 8.11 (6.48 - 10.71) | 8.91 (6.69 - 10.93) | 0.916 |
| Dome Volume, mm3 | 11.91 (7.46 - 19.42) | 13.30 (8.52 - 21.09) | 0.236 |
| VOR, mm | 1.45 (0.99 - 1.94) | 1.70 (1.25 - 2.14) | 0.059 |
| Hemodynamic parameters | |||
| FV, m/s | 0.18 (0.11 - 0.25) | 0.15 (0.09 - 0.26) | 0.568 |
| WSS, Pa | 4.95 (3.01 - 7.42) | 3.87 (2.51 - 8.12) | 0.723 |
| NWSS | 0.58 (0.41 - 0.83) | 0.59 (0.28 - 0.82) | 0.608 |
| OSI×10-3 | 5.13 (2.98 - 7.48) | 5.07 (2.57 - 7.28) | 0.985 |
| WSSG, Pa/mm | 6.33 (3.33 - 10.11) | 6.29 (3.01 - 8.43) | 0.730 |
| AFI | 0.975 (0.961 - 0.986) | 0.976 (0.960 - 0.986) | 0.989 |
| RRT, Pa-1 | 0.31 (0.17 - 0.52) | 0.36 (0.17 - 0.98) | 0.491 |
| FCR | 0.66 (0.54 - 0.89) | 0.61 (0.52 - 0.71) | 0.016 |
| LWSS | 0.83 (0.55 - 1.15) | 1.00 (0.59 - 1.13) | 0.371 |
| LSAR, 10-1 | 0.06 (0.00 - 0.43) | 0.16 (0.00 - 2.17) | 0.281 |
| HWSS | 5.44 (3.31 - 8.16) | 4.26 (2.77 - 8.93) | 0.723 |
| HSAR | 0.33 (0.28 - 0.35) | 0.28 (0.24 - 0.31) | < 0.001 |
| HSCR | 5.01 (4.29 - 6.51) | 6.39 (5.55 - 8.47) | < 0.001 |
Values are presented as median (IQR). *The p values are calculated using Brunner-Munzel tests between the stable and growth groups. AFI indicates aneurysm formation indicator; FCR, flow concentration ratio; FV, flow velocity at aneurysm dome; HSAR, high shear area ratio; HSCR, high shear concentration ratio; HWSS, high wall shear stress; LSAR, low shear area ratio; LWSS, low wall shear stress; NWSS, normalized wall shear stress; OSI, oscillatory shear index; RRT, relative residence time; VOR, volume-to-ostium area ratio; WSS, wall shear stress; and WSSG, wall shear stress gradient
Fig. 2.
Visualization of hemodynamic parameters in representative aneurysms from the stable group (A and B) and the growth group (C and D).
A and C show streamlines, with flow concentration ratios of 0.85 (A) and 0.53 (C). B and D show high shear area (HSA) and the magnitude of time-averaged wall shear stress within HSA. HSA ratio: 0.35 (B) and 0.30 (D), while high shear concentration ratio: 4.56 (B) and 6.89 (D). The ELAPSS scores are both 5 points.
Table 3.
ELAPSS score of aneurysms in the stable and growth groups
| Stable group (n=186) | Growth group (n=29) | p Value* | |
|---|---|---|---|
| Early SAH (0 - 1) | 0 (0 - 0) | 0 (0 - 0) | 0.251 |
| Location (0 - 5) | 0 (0 - 3) | 3 (0 - 3) | 0.349 |
| Age (0 - 6) | 1 (0 - 3) | 1 (0 - 3) | 0.763 |
| Population (0 - 7) | 1 (1 - 1) | 1 (1 - 1) | NaN |
| Size (0 - 22) | 4 (4 - 4) | 4 (4 - 4) | 0.319 |
| Shape (0 - 4) | 0 (0 - 0) | 0 (0 - 4) | 0.162 |
| Total score | 9 (6 - 11) | 9 (8 - 12) | 0.275 |
Values are presented as median (IQR). *The p values are calculated using Brunner-Munzel tests between the stable and growth groups. NaN indicates not a number. SAH indicates subarachnoid hemorrhage
ROC curve analysis
Among the hemodynamic parameters showing significant differences between the stable and growth groups in the univariate analysis, HSCR demonstrated the highest area under the ROC curve (AUC) value of 0.71 (95% confidence interval [CI] 0.61-0.81). The cut-off value was 5.38 with a sensitivity of 0.59 and a specificity of 0.83. The AUC values for FCR and HSAR were 0.62 (95% CI 0.53-0.72) and 0.69 (95% CI 0.60-0.79), respectively, with cut-off values of 0.83 and 0.33.
Discussion
In recent years, CFD has significantly advanced our comprehension of cerebral aneurysms, shedding light on their initiation, growth, and rupture.6-8,11,13-19) While CFD has unearthed novel insights, careful interpretation of its results is imperative. Given the influence of morphological variables,14) it was crucial to evaluate the aneurysm growth by aligning with the maximum diameter, as demonstrated in the previous report.8) Reassessing the ELAPSS score exclusively in the context of small aneurysms with a maximum diameter of 3 to 5 mm is valuable not only for enhancing growth prediction within its size range but also for evaluating hemodynamic parameters in small aneurysms without significant confounding influences.
Sonobe et al.23) indicated that risk factors for aneurysm growth in small unruptured cerebral aneurysms were 4 mm or larger size, women, multiple aneurysms, and smoking. In contrast, our study identified hypertension as a potential risk factor for aneurysm growth in small aneurysms. Additionally, FCR and HSAR were significantly lower and HSCR was significantly higher in the growth group, similar to previous reports in which the definition of growth was an increase in the size of more than 0.5 mm.8) Although morphological variables were not significantly different, the projection ratio and VOR, which have been reported in association with aneurysm growth and rupture tended to be higher in the growth group.8,12,24) ROC curve analyses highlighted that HSCR had the most diagnostically significant predictor of small aneurysm growth.
In this study, ELAPSS scores showed no significant differences in both individual scores and total scores between the stable and growth groups. Considering the morphological results, predicting the growth of small aneurysms based on the morphology and ELAPSS scores may be exceedingly challenging. The low inter-observer reproducibility in assessing irregular geometry was a critical factor in the ELAPSS score's limitations. Due to the lack of a strict definition of irregular shape, we assumed that it was extremely difficult to assess the slight differences in shape between small aneurysms. Improving inter-observer reproducibility, as suggested by Paritala et al.,25) would be essential for enhancing the predictive capability of the ELAPSS score. From this perspective, CFD measurement of hemodynamics appeared preferable to ELAPSS score for predicting small aneurysm growth.
Several limitations should be acknowledged in this study. Firstly, it is essential to consider the differences between the patient cohort in Backes et al.'s3) study on the ELAPSS score and the cohort used in this study. In the original study by Backes et al.,3) the patient cohort consisted of more than 1,500 individuals, which is over 7 times larger than the cohort in this study. Additionally, Backes et al.'s3) cohort included aneurysms with a maximum diameter exceeding 5 mm, showed no significant difference in the prevalence of hypertension between the stable and growth groups, and did not include data on smoking status. These differences in patient cohorts should be acknowledged as potential factors that may have influenced the results of this study. Secondly, the duration of patient observation in this study did not precisely align with the risk assessment timeline of the ELAPSS score, potentially impacting the evaluation of aneurysm growth. Therefore, we could not rule out the possibility of subsequent growth of aneurysms in the stable group. Thirdly, there are issues regarding the modality of evaluating aneurysms. In this study, only patients whose aneurysms were diagnosed by 3D CTA or 3D DSA were included for analysis using CFD. In other words, not all aneurysms were analyzed during the period of interest. Lastly, future studies would explore site-specific hemodynamic characteristics by increasing enrollment and evaluating the risk of growth based on aneurysm location.
Conclusions
This study suggests that, when limited to small aneurysms with a maximum diameter of 3 to 5 mm, CFD may be more effective than ELAPSS score in predicting their growth.
Conflicts of Interest Disclosure
All authors have no conflict of interest.
Supplementary Material
Acknowledgments
We thank the SMART Mie Group for their contribution to the data collection of unruptured cerebral aneurysms.
References
- 1).Inoue T, Shimizu H, Fujimura M, et al. Annual rupture risk of growing unruptured cerebral aneurysms detected by magnetic resonance angiography. J Neurosurg. 2012;117:20-5. [DOI] [PubMed] [Google Scholar]
- 2).Chmayssani M, Rebeiz JG, Rebeiz TJ, et al. Relationship of growth to aneurysm rupture in asymptomatic aneurysms≤ 7 mm: a systematic analysis of the literature. Neurosurgery. 2011;68:1164-71. [DOI] [PubMed] [Google Scholar]
- 3).Backes D, Rinkel GJ, Greving JP, et al. ELAPSS score for prediction of risk of growth of unruptured intracranial aneurysms. Neurology. 2017;88:1600-6. [DOI] [PubMed] [Google Scholar]
- 4).Investigators UJ. The natural course of unruptured cerebral aneurysms in a Japanese cohort. N Engl J Med. 2012;366:2474-82. [DOI] [PubMed] [Google Scholar]
- 5).Vindlacheruvu R, Mendelow A, Mitchell P. Risk-benefit analysis of the treatment of unruptured intracranial aneurysms. J Neurol Neurosurg Psychiatry. 2005;76:234-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6).Sforza DM, Kono K, Tateshima S, et al. Hemodynamics in growing and stable cerebral aneurysms. J Neurointerv Surg. 2016;8:407-12. [DOI] [PubMed] [Google Scholar]
- 7).Sugiyama S, Meng H, Funamoto K, et al. Hemodynamic analysis of growing intracranial aneurysms arising from a posterior inferior cerebellar artery. World Neurosurg. 2012;78:462-8. [DOI] [PubMed] [Google Scholar]
- 8).Tsuji M, Ishida F, Yasuda R, et al. Computational fluid dynamics for predicting the growth of small unruptured cerebral aneurysms. J Neurosurg. 2023;1:1-6. [DOI] [PubMed] [Google Scholar]
- 9).Ford MD, Alperin N, Lee SH, et al. Characterization of volumetric flow rate waveforms in the normal internal carotid and vertebral arteries. Physiol Meas. 2005;26:477. [DOI] [PubMed] [Google Scholar]
- 10).Murray CD. The physiological principle of minimum work applied to the angle of branching of arteries. J Gen Physiol. 1926;9:835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11).Tsuji M, Ishikawa T, Ishida F, et al. Stagnation and complex flow in ruptured cerebral aneurysms: a possible association with hemostatic pattern. J Neurosurg. 2017;126:1566-72. [DOI] [PubMed] [Google Scholar]
- 12).Yasuda R, Strother CM, Taki W, et al. Aneurysm volume-to-ostium area ratio: a parameter useful for discriminating the rupture status of intracranial aneurysms. Neurosurgery. 2011;68:310-8. [DOI] [PubMed] [Google Scholar]
- 13).Cebral JR, Mut F, Weir J, et al. Quantitative characterization of the hemodynamic environment in ruptured and unruptured brain aneurysms. AJNR Am J Neuroradiol. 2011;32:145-51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14).Fukazawa K, Ishida F, Umeda Y, et al. Using computational fluid dynamics analysis to characterize local hemodynamic features of middle cerebral artery aneurysm rupture points. World Neurosurg. 2015;83:80-6. [DOI] [PubMed] [Google Scholar]
- 15).Mantha A, Karmonik C, Benndorf G, et al. Hemodynamics in a cerebral artery before and after the formation of an aneurysm. AJNR Am J Neuroradiol. 2006;27:1113-8. [PMC free article] [PubMed] [Google Scholar]
- 16).Meng H, Wang Z, Hoi Y, et al. Complex hemodynamics at the apex of an arterial bifurcation induces vascular remodeling resembling cerebral aneurysm initiation. Stroke. 2007;38:1924-31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17).Miura Y, Ishida F, Umeda Y, et al. Low wall shear stress is independently associated with the rupture status of middle cerebral artery aneurysms. Stroke. 2013;44:519-21. [DOI] [PubMed] [Google Scholar]
- 18).Tanioka S, Ishida F, Kishimoto T, et al. Quantification of hemodynamic irregularity using oscillatory velocity index in the associations with the rupture status of cerebral aneurysms. J Neurointerv Surg. 2019;11:614-7. [DOI] [PubMed] [Google Scholar]
- 19).Tanioka S, Ishida F, Yamamoto A, et al. Machine learning classification of cerebral aneurysm rupture status with morphologic variables and hemodynamic parameters. Radiol Artif Intell. 2020;2:e190077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20).Backes D, Vergouwen MD, Velthuis BK, et al. Difference in aneurysm characteristics between ruptured and unruptured aneurysms in patients with multiple intracranial aneurysms. Stroke. 2014;45:1299-303. [DOI] [PubMed] [Google Scholar]
- 21).Brinjikji W, Zhu Y-Q, Lanzino G, et al. Risk factors for growth of intracranial aneurysms: a systematic review and meta-analysis. AJNR Am J Neuroradiol. 2016;37:615-20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22).Kanda Y. Investigation of the freely available easy-to-use software ‘EZR'for medical statistics. Bone Marrow Transplant. 2013;48:452-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23).Sonobe M, Yamazaki T, Yonekura M, et al. Small unruptured intracranial aneurysm verification study: SUAVe study, Japan. Stroke. 2010;41:1969-77. [DOI] [PubMed] [Google Scholar]
- 24).Ujiie H, Tamano Y, Sasaki K, et al. Is the aspect ratio a reliable index for predicting the rupture of a saccular aneurysm? Neurosurgery. 2001;48:495-503. [DOI] [PubMed] [Google Scholar]
- 25).Paritala PK, Anbananthan H, Hautaniemi J, et al. Reproducibility of the computational fluid dynamic analysis of a cerebral aneurysm monitored over a decade. Sci Rep. 2023;13:219. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.