Comparative analysis of the geometric morphology and local hemodynamics of small and medium intracranial aneurysms: a two-center retrospective study in a Chinese Han population

Introduction

Intracranial aneurysms (IAs) are pathological dilatations of intracranial arteries and have a prevalence of approximately 2–5% in the general population and an annual rupture rate of 0.7–1.9%. Once an IA ruptures, subarachnoid hemorrhage (SAH) and vasospasm occur, often leading to death and disability (1-3).

The size of the IA is an indicator used for assessing the risk of IA rupture (4,5), and a longitudinal study revealed that larger IAs are more likely to rupture (4). However, other studies have reported that 13% to 75% of ruptured IAs are very small (6,7) and therefore difficult to treat. Moreover, the definitions of small and large aneurysms differ. Both 7 and 10 mm are used as critical values (4,8,9). According to the International Study of Unruptured Intracranial Aneurysms (ISUIAs), IAs larger than 7 mm have a relatively high risk of rupture. In comparison, the risk of rupture is relatively low for IAs smaller than 7 mm (10). In natural history studies conducted in Japan (Unruptured Cerebral Aneurysm Study) and Finland, larger IAs were found to be more likely to rupture (10,11). Other research has revealed that a significant portion of ruptured IAs are small and located in the anterior communicating artery (ACoA) (10,12). In addition, small aneurysms (<7 mm) in other areas can also rupture; for example, in one study, among the 407 ruptured IAs of the middle cerebral artery (MCA), 29% were <7 mm in size at the time of rupture (13).

Greving et al. (8) introduced a PHASES (population, hypertension, age, aneurysm size, SAH history, and aneurysm location) score to predict IA rupture on the basis of aggregated data analysis. In their study, they found age, hypertension, SAH history, aneurysm size, and aneurysm location to be independent predictors of aneurysm rupture.

Compared with populations in countries outside of North America and Finland, Finnish people are at 3.6 times greater risk of aneurysm rupture, whereas Japanese people are at 2.8 times greater risk. However, there are no reports of single small- or medium-sized aneurysms in the Chinese population.

According to clinical evidence, small IAs and large IAs have different pathophysiological manifestations (14). An intraoperative study revealed that smaller IAs (7–10 mm) often have thin, translucent walls, whereas larger IAs (12–20 mm) typically have thick, irregular, and white walls (15). Given that the size of an aneurysm may not fully reflect its risk of rupture, we hypothesized that small and large IAs may have different morphological, hemodynamic, and clinical characteristics. In this study, we investigated the local hemodynamics and clinical and geometric morphology of small- and medium-sized (<15 mm) single IAs in a Han Chinese population to clarify their differences and provide guidelines for clinical management in this population. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-943/rc).

Methods

Study population and case selection

Between January 1, 2016, and May 31, 2022, we initially identified 779 consecutive patients with documented IAs based on cranial computed tomography angiography (CTA) or digital subtraction angiography (DSA) imaging from the radiology information systems of Tianjin Medical University General Hospital and Tianjin Fifth Central Hospital.

After removal of duplicates, two board-certified neuroradiologists independently reviewed the imaging data and corresponding clinical documentation. Patients were excluded based on the following criteria: (I) presence of fusiform, traumatic, mycotic, or postoperative pseudoaneurysms; (II) aneurysm diameter ≥15 mm; (III) inadequate image quality precluding three-dimensional (3D) reconstruction or hemodynamic analysis; and (IV) incomplete clinical information.

For cohort classification, the remaining cases were categorized as ruptured IAs (RIAs) if aneurysmal SAH was confirmed by computed tomography (CT), magnetic resonance imaging (MRI), or intraoperative findings within 24 hours of the index imaging. Cases without evidence of rupture were classified as unruptured IAs (UIAs).

Because all eligible cases within the defined study period were included regardless of how the aneurysm was discovered (e.g., incidentally, during routine screening or in symptomatic patients), our sampling was consecutive rather than convenience-based. This approach minimized selection bias and ensured that the UIAs and RIAs originated from the same underlying population, time frame, and institutional referral pathways. The detailed patient enrollment process is shown in Figure 1.

Figure 1 Flowchart of patient selection. UIAs were defined as aneurysms without evidence of rupture on admission and no SAH on CT or surgical confirmation. RIAs were diagnosed based on acute clinical presentation with SAH confirmed by imaging or surgery. 3D, three-dimensional; CFD, computational fluid dynamics; CT, computed tomography; CTA, computed tomography angiography; DSA, digital subtraction angiography; IA, intracranial aneurysm; RIA, ruptured intracranial aneurysm; SAH, subarachnoid hemorrhage; UIA, unruptured intracranial aneurysm.

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and was approved by the Ethics Committee of Tianjin Medical University General Hospital (approval No. IRB2022-YX-175-01) and Tianjin Fifth Central Hospital (approval No. WZX-EC-KY2024039). Given the retrospective nature of the study and the use of de-identified data, the requirement for informed consent was formally waived by the respective ethics committees. All data were anonymized prior to analysis to protect patient confidentiality.

3D models and definition of geometric parameters

Mimics Research 21.0 (Materialise Inc., Leuven, Belgium) software was used to create a 3D model of the IA on the basis of CTA and DSA images stored in Digital Imaging and Communications in Medicine (DICOM) format (Figure 2), and the geometric morphology of the aneurysm was measured after reconstruction (Figure 3A). Representative models illustrating inflow direction, bleb location, daughter sac, and regional division into neck, body, and dome are shown in Figure 3B-3D. All aneurysm models were reconstructed from multidetector CTA (slice thickness 0.5–0.625 mm) or 3D rotational DSA (voxel size 0.10–0.20 mm). Magnetic resonance angiography (MRA) was not used due to its lower spatial resolution and flow-related artifacts, which might have biased the morphological assessment. Table S1 provides definitions of the morphological parameters. We referred to Salimi’s concept of dividing IAs into three regions (non-software partitioning method): the neck, body, and dome (16).

Figure 2 Reconstruction of blood vessels in three dimensions. (A-D) Reconstruction of CTA DICOM files via Mimics Research 21.0 software. (E-H) Reconstruction of DICOM format files for DSA via Mimics Research 21.0 software. 3D, three-dimensional; CTA, computed tomography angiography; DSA, digital subtraction angiography; DICOM, Digital Imaging and Communications in Medicine.

Figure 3 Geometric measurements and regional division of IAs. (A) Schematic diagram of geometric parameters. H = height; L = length; Wneck = neck width; α (red curved line) = incidence angle; β (black curved line) = aneurysm angle. The black arrow (flow) indicates blood flow direction. The parent-vessel mean diameter (Dv) is obtained as follows: for side-wall IAs, it is the average of D1 (diameter adjacent to the neck) and 1.5× D2 (1.5 vessel diameters proximal to the neck); for bifurcation IAs, it is the average diameter of the parent vessel and its daughter branches. (B) The black arrow (flow) shows inflow direction. The black arrow (bleb) indicates a bleb on the dome. The labels ‘neck’, ‘body’, and ‘dome’ are in black; region boundaries are indicated by black dashed lines. (C) IA with identical annotations: the black arrows (flow and bleb) indicate inflow and bleb, respectively. The red text indicates the neck, body, and dome regions. (D) IA with a daughter sac. The black arrows (flow and daughter sac) indicate the main flow stream and the secondary sac, respectively. The red line segment indicates the neckline (neck plane). IAs, intracranial aneurysms.

3D model processing and finite-volume mesh generation

The images of the 3D models were exported from Mimics Research 21.0 software in STL format and used for clipping, further smoothing, and end cutting in 3-matic Research 13.0 (×64; Materialise Inc.). We imported the images of the processed 3D model into ICEM CFD 2021 R2 software (Ansys Inc., Canonsburg, PA, USA). This model used unstructured tetrahedral grids with a maximum grid diameter of 0.4 mm (approximately 800,000–1,200,000 finite grids). Mesh independence was confirmed in four representative cases by comparing coarse, medium, and fine meshes; the key hemodynamic differences between medium and fine meshes were <2%. To assess segmentation-related uncertainty, a random subset of 10 aneurysms was resegmented with a ±20 Hounsfield unit (HU) threshold and a ±0.2 mm smoothing radius. Changes in key geometric [aspect ratio (AR) and size ratio (SR)] and hemodynamic [time-averaged wall shear stress of the neck (TAWSS_neck), and oscillatory shear index of the neck (OSI_neck)] parameters were all minor (<5%), indicating limited impact on our results.

Control equations, boundary conditions, and data manipulation

In this study, blood was assumed to be a Newtonian fluid, and energy transfer and gravity were not considered. The basic equation for controlling flow is the 3D viscous incompressible equation, in which the wall of the IA is assumed to be rigid, without sliding boundaries, and the effects of wall thickness and elasticity on hemodynamic parameters are ignored. The blood flow conditions at the entrance of the vessel were set on the basis of the pulsatile flow velocity measured by transcranial Doppler (TCD) ultrasound [the average flow velocity at the entrance of the internal carotid artery (ICA) in patients aged 40 to 60 years was 0.92±0.10 m/s according to TCD] and the outlet pressure condition (no stress; i.e., the static pressure value was 0). For each case, inflow was scaled to parent artery diameter (Qcalc = A × V_ref), and outlets were set to pressure (0 Pa) with flow split equally or by Q∝D³.

In this study, Fluent 2021 R2 finite-volume analysis software (Ansys Inc.) was used to construct the medical subject headings (MeSH) vascular model for hemodynamic analysis, and the Navier-Stokes equation was used to govern the simulation process. During each cardiac cycle, 800 steps were simulated (0.01 seconds per step), with three cardiac cycles being simulated and data being exported from the most recent cardiac cycle. The exported calculation results were saved as CDAT files for analysis, and the obtained image data were postprocessed via CFD Post 2021 R2 software (Ansys Inc.). Additionally, to assess shear-thinning effects, four representative aneurysms were re-simulated via the Carreau-Yasuda model under identical settings; the results did not materially change our conclusions.

Statistical analysis

The size and morphological characteristics of the IA were analyzed. One-way analysis was performed for the above-mentioned parameters, the Mann-Whitney test (z±S) was applied for measurement data, and the Chi-squared test (χ²) was used for count data. The t-test was used for normally distributed data, and the Wilcoxon rank-sum test was used for nonnormally distributed data. An analysis of the receiver operating characteristic (ROC) curve was conducted for the size and risk of rupture of the IA to determine the area under the curve (AUC), and the Youden index was used to determine the optimal critical value (P<0.05). We measured the local area parameters by randomly selecting multiple points near the neck, body, and dome of the aneurysm and then taking the mean of these measurements to represent the hemodynamic parameters of their local area. Statistical analyses were performed with Excel 2010 (Microsoft Corp., Redmond, WA, USA) and SPSS l9.0 (IBM Corp., Armonk, NY, USA).

Results

A total of 214 small- and medium-sized (<15 mm) aneurysms were included in the analysis (MRA images were not used for morphology reconstruction; see Methods for the rationale). There were 93 men and 121 women, with a mean age of 59.90±11.76 years: 56.24±12.96 years for men (range, 16–90 years) and 62.71±9.85 years for women (range, 32–85 years). The rupture rate was higher for medium-sized aneurysms (67.4%) than for small aneurysms (44.7%) (P=0.001). Diabetes (P=0.016), coronary heart disease (P=0.017), and atherosclerosis (P=0.003) were associated with a higher incidence of small aneurysms. The distribution of the location of small- and medium-sized aneurysms significantly differed (P<0.001). Small aneurysms were most commonly located in the ICA, anterior cerebral artery (ACA), and sidewall areas, and medium-sized aneurysms were most commonly located in the ACoA, posterior communicating artery (PCoA), MCA, and arterial bifurcation areas. Compared with smaller aneurysms, medium aneurysms had higher AR values, SR values, bottleneck ratio (BN) values, height-to-width ratio (HWR) values, and incidence angles (P<0.001) (Table 1). Mesh convergence analysis showed ≤1.8 % and ≤1.6 % differences in TAWSS_Neck and OSI_Dome between medium and fine meshes, respectively (and <5% and 4.2% between coarse and fine meshes, respectively), confirming that the medium mesh density is sufficient for hemodynamic accuracy.

Patients with small aneurysms were more likely to be young at the time of rupture (P=0.016). Among small aneurysms, those located at the ACoA, PCoA, and MCA, as well as at bifurcation sites, were more likely to rupture (P<0.001). Moreover, within the small aneurysm group, higher SR (P=0.044), incidence angle (P=0.011), and aneurysm angle (P=0.007) were also associated with rupture. In contrast, among the medium-sized aneurysms, irregular shape (P=0.007) and bifurcation location (P<0.001) were more associated with rupture (Table 2).

According to the ROC curve analysis, IA size (L>3.87 mm) was associated with a higher risk of rupture, indicating that compared with small-sized IAs, medium-sized IAs are more prone to rupture [AUC =0.613; 95% confidence interval (CI): 0.535–0.692; P=0.015] (Figure 4). The OSI (Pneck=0.033; Pdome=0.003) and mean relative residence time (RRT) (Pneck=0.028, Pdome=0.003) at the neck and dome sites of small- and medium-sized aneurysms were significantly different, and the values for medium-sized aneurysms were lower. The dome sites of small- and medium-sized aneurysms significantly differed in terms of TAWSS (P=0.04), mean time-averaged wall shear stress gradient (TAWSSG) (P=0.04), and aneurysm formation indicator (AFI) parameters (P=0.031), with medium-sized aneurysms having smaller TAWSS and TAWSSG values and larger AFI values. The neck and body of medium-sized IAs differed significantly in OSI (P=0.028) and RRT (P=0.002), whereas there was no significant difference in small IAs (Table 3). Moreover, for medium-sized aneurysms, the neck and body differed significantly in terms of OSI (P=0.028) and RRT (P=0.002) (Table 4). Examples of hemodynamic parameters for RIAs and UIAs in small- and medium-sized IAs are shown in Figures 5,6 and Figure S1. Sensitivity analysis confirmed that the hemodynamic results were robust to reasonable variations in inflow.

Figure 4 ROC curve analysis of the correlation between the size and rupture of an IA. AUC, area under the curve; CI, confidence interval; IA, intracranial aneurysm; ROC, receiver operating characteristic.

Figure 5 The hemodynamics of ruptured and unruptured small IAs. (A-F) Ruptured MCA IA with a length of 1.6 mm. (A) Three-dimensional reconstruction model. (B) TAWSS distribution. (C) TAWSSG distribution. (D) OSI distribution. (E) Mean RRT distribution. (F) AFI. (G-L) Unruptured MCA IA with a length of 1.46 mm. (G) Three-dimensional reconstruction model. (H) TAWSS. (I) TAWSSG. (J) OSI. (K) RRT. (L) AFI. AFI, aneurysm formation indicator; IA, intracranial aneurysm; MCA, middle cerebral artery; OSI, oscillatory shear index; RRT, relative residence time; TAWSS, time-averaged wall shear stress; TAWSSG, time-averaged wall shear stress gradient.

Figure 6 The hemodynamics of ruptured and unruptured medium-sized IAs. (A-F) Ruptured MCA IA with a length of 6.08 mm. (A) Three-dimensional reconstruction model. (B) TAWSS distribution. (C) TAWSSG distribution. (D) OSI distribution. (E) Mean RRT distribution. (F) AFI. (G-L) Unruptured MCA IA with a length of 5.56 mm. (G) Three-dimensional reconstruction model. (H) TAWSS. (I) TAWSSG. (J) OSI. (K) RRT. (L) AFI. AFI, aneurysm formation indicator; IA, intracranial aneurysm; MCA, middle cerebral artery; OSI, oscillatory shear index; RRT, relative residence time; TAWSS, time-averaged wall shear stress; TAWSSG, time-averaged wall shear stress gradient.

Discussion

Relationship between clinical factors and IA rupture

Age has traditionally been considered a significant risk factor for IA rupture due to age-related vascular degeneration and cerebral atherosclerosis (3). However, recent epidemiological data from 8,144 aneurysmal SAH cases in mainland China revealed a lower incidence in the older adults compared with middle-aged individuals (17), suggesting that age-related rupture risk may vary by population. In our study, ruptured small IAs occurred more often in younger patients (mean age 55.71 years) than did medium-sized ruptured IAs (mean age 59.41 years), aligning with evidence that younger age independently predicts rupture risk in small aneurysms due to more dynamic hemodynamics, higher inflammation, and fewer comorbidities. A multicenter cohort study in Japan similarly reported a 5.23-fold higher rupture risk in patients aged <50 years for IAs <5 mm (18). These findings, consistent with those of Suzuki et al. (19) and Takao et al. (7), underscore the need for vigilance in younger individuals, especially for high-risk locations such as the ACoA and PCoA. Zhang et al. reported that patients with ruptured PCoA aneurysms were younger than those with unruptured (56.7 ± 15 vs. 61.8 ± 13 years) (20). Collectively, our results confirm that younger age is associated with increased rupture risk, consistent with previous research.

Other reported clinical risk factors include smoking, hypertension, and diabetes. Smoking is an established risk factor for UIAs and SAH, with women also at increased risk. Müller et al. reported that systolic blood pressure is associated with IA rupture but not with formation (21). Diabetes, although linked to atherosclerosis, has been associated with a lower risk of SAH and aneurysms in (22,23). In our study, hypertension, smoking habit, and related factors did not differ significantly between groups. However, diabetes and related comorbidities were common and associated with lower rupture rates, consistent with earlier findings (22,23). Although type 2 diabetes mellitus appears to be negatively correlated with rupture risk, the related underlying mechanisms remain unclear, but may possibly involve better medical management, lifestyle modifications, or confounding effects of other diabetes-related complications.

Relationship between IA size and rupture

The PHASES (Population, Hypertension, Age, Body Size, Early Subarachnoid Hemorrhage, and IA Site) study (24), a longitudinal study including the ISUIA and the International Study of Unruptured Cerebral Aneurysms (ISUCAS) (9), revealed that the estimated risk of rupture was correlated with aneurysm size (25).

Murayama et al. reported that the size and location of the aneurysm and a history of SAH are important independent predictors of rupture. Compared with aneurysms measuring 2–4 mm in length, those measuring 5 mm in length are significantly more likely to rupture (26). These results were also confirmed by stratified analysis of IAs <4, ≥4, and <15 mm in size.

IA size is currently an important geometric morphological parameter for predicting the risk of IA rupture. The ISUIA reported that the annual rupture rate was 0.05% for IAs <10 mm and 1% for IAs ≥10 mm. In the ISUIA, an IA size of 10 mm was used as the threshold for rupture risk, which is highly controversial, and numerous other clinical studies have revealed that most ruptured IAs are <10 mm in size (7,8). In our study, the rate of medium-sized aneurysm rupture (67.4%) was higher than that of small-sized aneurysm rupture (44.7%) (P=0.001). Our ROC curve analysis identified 3.87 mm as the optimal threshold for rupture prediction (AUC =0.613), with a sensitivity of 0.71 and a specificity of 0.55, indicating that more than 40% of ruptured aneurysms were smaller than 5 mm, even after location and SR were accounted for. Although the overall rupture rate increases with aneurysm size, numerous studies, including ours, have found a considerable proportion of ruptured aneurysms are smaller than 5 mm. For instance, a Chinese single-center study of 1256 cases of sporadic ruptured cerebral aneurysms reported that 38.6% of ruptured IAs were smaller than 5 mm in diameter (27), which is consistent with our findings.

In a large multinational follow-up study, the risk of rupture was lower for IAs smaller than 7 mm. However, observational data indicate that a substantial proportion of small IAs rupture, especially when located in the ACoA (10,12). A follow-up study in Japan revealed that IAs located in the anterior or posterior communicating arteries are at high risk of rupture, regardless of size (25). According to these natural history studies, although the average risk of rupture of small IAs is low, a portion of small IAs are susceptible to rupture (10,25) despite their small size (10,24). Although a larger aneurysm is at higher risk of rupture, the optimal size for treatment remains unclear (26).

Whether to treat or observe aneurysms larger than 5 mm and smaller than 7 mm has not been determined. The risk associated with the natural course of the aneurysm, the patient’s life expectancy, and the risk of treatment must be considered in this decision. The risk increases 2.7-fold for aneurysms between 7 and 10 mm and 14.3-fold for aneurysms greater than 20 mm. According to the Unruptured Cerebral Aneurysm Study (UCAS) in Japan, the risk of rupture of aneurysms measuring 5–6 mm is similar to that of aneurysms measuring 3–4 mm (25,26). In contrast, in our study, the rupture rate of aneurysms larger than 4 mm was higher. The IA size does not represent the geometric morphological characteristics of the IA, and the risk of rupture cannot be fully assessed through IA size alone. In short, size is an important but not exclusive determinant of rupture. Geometry-derived indices and local hemodynamic patterns provide critical complementary information for risk stratification. Other geometric morphological characteristics must be further explored for predicting the risk of IA rupture.

Relationship between IA shape and rupture

In a study by Abboud et al., morphology was an independent predictor of IA rupture, and the associated risk changed with morphology. Previous studies evaluated RIAs and UIAs, classified their morphology into regular and irregular, and reported that irregular shapes strongly correlated with RIAs (28,29). In our study, irregular shape was observed in 58.6% of ruptured aneurysms and in 27.3% of unruptured aneurysms [odds ratio (OR) 3.7; 95% CI: 2.1–6.4], especially at bifurcation sites such as the ACoA and PCoA. Zhang et al. reported that AR values were significantly higher for RIAs than for UIAs and that RIAs tended to have irregularly shaped bifurcations (28,30).

In Suzuki et al.’s study (19), younger age, multiple aneurysms, bifurcation location, presence of a bleb, longer length, and a lower pressure loss coefficient were identified as risk factors for the rupture of small IAs. Moreover, no significant association was found between the aneurysm location and AR value and rupture risk of small aneurysms. According to Miyata et al., a sharper bifurcation angle and an inclination angle between the M1 and M2 arteries are associated with the rupture of IAs of the MCA (31). Our study revealed that aneurysms at the bifurcation of the ACoA and PCoA were more likely to rupture and that a high incidence of aneurysm angles was associated with an increased risk of rupture. The risk of rupture is greater for medium-sized aneurysms with irregular shapes and those located at bifurcations.

In Lindgren et al.’s nonselective population registry study, regardless of patient background or IA size, irregularly shaped IAs at various sites were more likely to rupture than were regularly shaped IAs (10). In addition, despite the clear relationship between IA size and rupture rate, the relationship between irregular shape and rupture was significantly stronger than that between irregular shape and other known rupture risk factors. According to the findings of our study, compared with small- and medium-sized IAs, irregularly shaped IAs with blebs are more susceptible to rupture.

Relationships between AR and SR values and IA rupture

The larger the AR value is, the greater the risk of IA rupture, which is generally considered to be correlated with intra-aneurysmal blood flow in the IA. However, Merritt et al. (32) emphasized that inconsistent definitions of aneurysm size and morphology (including aspect ratio and dome-to-neck ratio) across studies complicate comparisons of hemodynamic results; therefore, reported relationships between AR, intra-aneurysmal flow, and thrombosis thresholds should be interpreted with caution. In our study, the AR value of medium-sized IAs (0.73±0.26) was greater than that of small IAs (1.21±0.46) (P<0.001). However, when the RIA group was compared with the UIA group for small IAs, there was no statistically significant difference in the AR value. Similar results were also obtained for medium-sized IAs.

SR, first reported by Dhar et al., is a parameter related to the size of the IA and the IA-bearing artery (4). Dhar et al. found a significant difference (P<0.05) in the mean diameter of aneurysm-carrying arteries between ruptured and unruptured IAs (4). After these low IA rupture rates were excluded, the mean values of the carrier artery diameters were again compared between the two groups, and the difference was not statistically significant (P=0.125) (33). In our study, there was no statistically significant difference in the diameter of the parent aneurysm between small- and medium-sized IAs (P=0.748). For medium-sized IAs, the diameter of the carrying artery in the RIA group (3.27±0.89) was smaller than that in the UIA group (3.64±0.89) (P=0.027).

Our data suggest that SR >0.862 may serve as a reliable threshold for predicting rupture risk in both small and medium IAs, particularly those located at bifurcation points such as the ACoA and PCoA. Although AR did not show an independent association with rupture after adjustment, we found that for both small and medium IAs, SR >0.862 nearly doubled the odds of rupture (OR 1.9; 95% CI: 1.3–2.8). This is in agreement with previous findings (34-37) that higher SR values are associated with rupture and refines earlier estimates of a SR ≥1.21 threshold proposed by Duan et al. (34). Although we did not stratify by location, our ROC analysis supports the utility of SR as a geometry-derived marker of rupture susceptibility in Chinese patients. In terms of mechanism, higher SR may increase the size of the flow impingement zone and steepen spatial WSS gradients, which can promote oscillatory flow and localized endothelial injury.

Hemodynamics and potential mechanisms

Previous studies, including longitudinal and cross-sectional studies, have investigated IAs without dividing them into two categories on the basis of their size. However, small and large IAs may present different phenotypes or pathological types (14,38). Varble et al. (38) stratified and analyzed two IA groups (<5 and >5 mm) in their study. They reported that larger IAs (≥5 mm) are more likely irregularly shaped and have lower WSS values, whereas smaller IAs (<5 mm) are more spherical and have higher WSS values. According to our study, compared with small aneurysms, medium-sized aneurysms were irregularly shaped and had lower TAWSS and TAWSSG values at the dome site. Specifically, for medium IAs, as compared to small IAs, the dome-level TAWSS was 23% lower, the OSI was 31% higher, and the RRT was 28% higher; meanwhile, for the neck, the TAWSS was 18% lower, while the OSI was 24% higher. Hemodynamic differences were robust to reasonable inflow and rheology variations, confirming the stability of our conclusions. Carreau-Yasuda testing showed ≤5% differences from Newtonian testing, and thus rheology choice does not appear to alter rupture risk.

According to Meng et al. (14), a low wall shear stress value may lead to large aneurysms and atherosclerotic aneurysms through inflammatory reactions. In contrast, a high wall shear stress value may lead to the growth and rupture of small aneurysms or thrombi through mural cell pathways. Low-shear and high-OSI environments promote endothelial dysfunction and inflammatory cell infiltration, consistent with the inflammation-centered mechanism proposed by Meng et al. (14). Conversely, regions of higher shear and high SR may trigger mural cell activation and extracellular matrix remodeling, as suggested by Varble et al. (38).

The vascular WSS and its spatial gradient may vary according to IA location, whereas aneurysms in non-bifurcated areas are considered to have atherosclerotic aneurysm characteristics (14,22).

In our study, small aneurysms were more likely to be located in the ICA, ACA, and sidewall areas. In contrast, medium-sized aneurysms were more likely to be located in the ACoA, PCoA, MCA, and arterial bifurcation areas. The TAWSS, TAWSSG, and AFI values significantly differed between small- and medium-sized aneurysms at the dome site. Therefore, it is necessary to pay greater attention to the dome area. This may be indirectly explained by differences in their anatomical location, geometric shape, and local hemodynamics, as they rupture in different pathways.

Furthermore, in inflammatory cell-mediated atherosclerosis, the cells are large and irregularly shaped, whereas, in mural cell-mediated atherosclerosis, the cells are small and smooth (14,38). According to Varble et al.’s study, a low wall shear stress value indicates the rupture of a large IA. This may suggest that inflammatory cell-mediated pathways are predominant in these IAs. Meanwhile, a high OSI value indicates the rupture of a small IA (38). In our study, compared with large IAs, small IAs had a relatively larger surface area and higher OSI values. In addition to hemodynamics, small and large IAs differed in location and, and as in previous studies, large IAs of the ACA and PCoA tended to rupture relatively frequently (8,38). In the neck and dome sites, the OSI and RRT values differed significantly between small- and medium-sized aneurysms, indicating that small aneurysms have higher values. Large sample, prospective studies are needed to better ascertain the geometric morphology, hemodynamics, and wall characteristics of IAs.

Flow coupling geometry

Our findings support the following cascade model: greater SR and irregular shape → enlarged flow impingement zone → steeper wall shear stress gradients and elevated OSI/RRT → increased wall inflammation and remodeling. This mechanistic link helps explain why some small aneurysms rupture while others remain stable and underscores the importance of integrating geometric and hemodynamic metrics for risk prediction in clinical practice.

Limitations

This study involved several limitations which should be acknowledged. First, our retrospective, cross-sectional design across two tertiary centers limits our ability to infer causal relationships between the identified clinical, geometric, and hemodynamic characteristics and aneurysm rupture risk. Rather, these findings demonstrate associations and are not definitive predictive evidence, and thus prospective longitudinal studies are required for validation.

Second, to minimize potential sampling bias, we employed a consecutive inclusion strategy across defined periods at both centers, selecting RIAs and UIAs based on identical criteria and sources. UIAs included incidental findings from imaging performed for other medical conditions, reflecting a broad clinical spectrum. This approach strengthened the internal validity and generalizability of our comparisons between groups.

Finally, due to our cross-sectional analysis, we could not determine whether individual small aneurysms might progress to medium-sized aneurysms over time. Accumulating longitudinal data indicate that only a subset of small aneurysms exhibits measurable growth, while others remain stable or rupture prior to enlargement (25,39,40). Our observed hemodynamic differences—higher OSI and RRT in small aneurysms compared to lower TAWSS and larger inflow angles in medium aneurysms—support distinct biological trajectories: (I) rapid rupture in small aneurysms associated with high shear oscillation and (II) chronic remodeling in medium aneurysms characterized by progressive neck widening and flow deceleration. Thus, our findings do not imply that small aneurysms will necessarily remain small but rather suggest that aneurysm growth is not a prerequisite for rupture. Future prospective studies with serial imaging are essential to confirming these hypotheses. Fourth, although semiautomated segmentation may introduce some user-dependent variability, our dedicated sensitivity analysis demonstrated that all key geometric and hemodynamic parameters deviated less than 5% under plausible segmentation parameter changes, supporting the robustness of our conclusions. Fifth, we did not include MRA-based reconstructions due to lower spatial resolution and greater artifact risk compared to CTA and DSA, which could have led to bias in the morphology and CFD results.

Conclusions

Small and medium aneurysms exhibit distinct geometric and hemodynamic characteristics associated with their rupture status. However, given the retrospective cross-sectional nature of this study, these findings represent associative rather than causal evidence, and prospective longitudinal studies are needed to confirm their role in predicting rupture risk.

Acknowledgments

This study was sponsored by the National Key Specialty Construction Project (2023283 to X.C.).

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-943/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-943/dss

Funding: This study was sponsored by Tianjin Health Research Project (grant No. TJWJ2023QN102 to X.C.), the Peking University Binhai Hospital Research Innovation and Transformation Fund (grant No. 2024-CX-01 to X.C.) and the projects of the Natural Science Foundation of Tianjin, China (No. 21JCQNJC00350).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-943/coif). X.C. reports this study was sponsored by Tianjin Health Research Project (grant No. TJWJ2023QN102), the Peking University Binhai Hospital Research Innovation and transformation Fund (grant No. 2024-CX-01), and the National Key Specialty Construction Project (grant No. 2023283). The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and approved by the Ethics Committee of Tianjin Medical University General Hospital (approval No. IRB2022-YX-175-01) and Tianjin Fifth Central Hospital (approval No. WZX-EC-KY2024039). Given the retrospective nature of the study and the use of de-identified data, the requirement for informed consent was formally waived by the respective ethics committees. All data were anonymized prior to analysis to protect patient confidentiality.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Neulen A, Pantel T, König J, Brockmann MA, Ringel F, Kantelhardt SR. Comparison of Unruptured Intracranial Aneurysm Treatment Score and PHASES Score in Subarachnoid Hemorrhage Patients With Multiple Intracranial Aneurysms. Front Neurol 2021;12:616497.
2. Lee WK, Oh CW, Lee H, Lee KS, Park H. Factors influencing the incidence and treatment of intracranial aneurysm and subarachnoid hemorrhage: time trends and socioeconomic disparities under an universal healthcare system. J Neurointerv Surg 2019;11:159-65.
3. Byoun HS, Huh W, Oh CW, Bang JS, Hwang G, Kwon OK. Natural History of Unruptured Intracranial Aneurysms : A Retrospective Single Center Analysis. J Korean Neurosurg Soc 2016;59:11-6.
4. Dhar S, Tremmel M, Mocco J, Kim M, Yamamoto J, Siddiqui AH, Hopkins LN, Meng H. Morphology parameters for intracranial aneurysm rupture risk assessment. Neurosurgery 2008;63:185-96; discussion 196-7.
5. Antiga L, Piccinelli M, Botti L, Ene-Iordache B, Remuzzi A, Steinman DA. An image-based modeling framework for patient-specific computational hemodynamics. Med Biol Eng Comput 2008;46:1097-112.
6. Xiang J, Natarajan SK, Tremmel M, Ma D, Mocco J, Hopkins LN, Siddiqui AH, Levy EI, Meng H. Hemodynamic-morphologic discriminants for intracranial aneurysm rupture. Stroke 2011;42:144-52.
7. Takao H, Murayama Y, Otsuka S, Qian Y, Mohamed A, Masuda S, Yamamoto M, Abe T. Hemodynamic differences between unruptured and ruptured intracranial aneurysms during observation. Stroke 2012;43:1436-9.
8. Greving JP, Wermer MJ, Brown RD Jr, Morita A, Juvela S, Yonekura M, Ishibashi T, Torner JC, Nakayama T, Rinkel GJ, Algra A. Development of the PHASES score for prediction of risk of rupture of intracranial aneurysms: a pooled analysis of six prospective cohort studies. Lancet Neurol 2014;13:59-66.
9. Wiebers DO, Whisnant JP, Huston J 3rd, Meissner I, Brown RD Jr, Piepgras DG, Forbes GS, Thielen K, Nichols D, O'Fallon WM, Peacock J, Jaeger L, Kassell NF, Kongable-Beckman GL, Torner JC. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet 2003;362:103-10.
10. Lindgren AE, Koivisto T, Björkman J, von Und Zu Fraunberg M, Helin K, Jääskeläinen JE, Frösen J. Irregular Shape of Intracranial Aneurysm Indicates Rupture Risk Irrespective of Size in a Population-Based Cohort. Stroke 2016;47:1219-26.
11. Korja M, Lehto H, Juvela S. Lifelong rupture risk of intracranial aneurysms depends on risk factors: a prospective Finnish cohort study. Stroke 2014;45:1958-63.
12. Bijlenga P, Ebeling C, Jaegersberg M, Summers P, Rogers A, Waterworth A, Iavindrasana J, Macho J, Pereira VM, Bukovics P, Vivas E, Sturkenboom MC, Wright J, Friedrich CM, Frangi A, Byrne J, Schaller K, Rufenacht D. Risk of rupture of small anterior communicating artery aneurysms is similar to posterior circulation aneurysms. Stroke 2013;44:3018-26.
13. Elsharkawy A, Lehečka M, Niemelä M, Billon-Grand R, Lehto H, Kivisaari R, Hernesniemi J. A new, more accurate classification of middle cerebral artery aneurysms: computed tomography angiographic study of 1,009 consecutive cases with 1,309 middle cerebral artery aneurysms. Neurosurgery 2013;73:94-102; discussion 102.
14. Meng H, Tutino VM, Xiang J, Siddiqui A. High WSS or low WSS? Complex interactions of hemodynamics with intracranial aneurysm initiation, growth, and rupture: toward a unifying hypothesis. AJNR Am J Neuroradiol 2014;35:1254-62.
15. Kadasi LM, Dent WC, Malek AM. Cerebral aneurysm wall thickness analysis using intraoperative microscopy: effect of size and gender on thin translucent regions. J Neurointerv Surg 2013;5:201-6.
16. Salimi Ashkezari SF, Mut F, Chung BJ, Robertson AM, Cebral JR. Hemodynamic conditions that favor bleb formation in cerebral aneurysms. J Neurointerv Surg 2021;13:231-6.
17. Wáng YXJ, Zhang L, Zhao L, He J, Zeng XJ, Liu H, et al. Elderly population have a decreased aneurysmal subarachnoid hemorrhage incidence rate than Middle aged population: a descriptive analysis of 8,144 cases in mainland China. Br J Neurosurg 2018;32:165-71.
18. Jing L, Fan J, Wang Y, Li H, Wang S, Yang X, Zhang Y. Morphologic and Hemodynamic Analysis in the Patients with Multiple Intracranial Aneurysms: Ruptured versus Unruptured. PLoS One 2015;10:e0132494.
19. Suzuki T, Takao H, Rapaka S, Fujimura S, Ioan Nita C, Uchiyama Y, Ohno H, Otani K, Dahmani C, Mihalef V, Sharma P, Mohamed A, Redel T, Ishibashi T, Yamamoto M, Murayama Y. Rupture Risk of Small Unruptured Intracranial Aneurysms in Japanese Adults. Stroke 2020;51:641-3.
20. Zhang J, Can A, Lai PMR, Mukundan S Jr, Castro VM, Dligach D, Finan S, Yu S, Gainer VS, Shadick NA, Savova G, Murphy SN, Cai T, Weiss ST, Du R. Age and morphology of posterior communicating artery aneurysms. Sci Rep. 2020;10:11545.
21. Müller TB, Vik A, Romundstad PR, Sandvei MS. Risk Factors for Unruptured Intracranial Aneurysms and Subarachnoid Hemorrhage in a Prospective Population-Based Study. Stroke 2019;50:2952-5.
22. Cho BH, Kim JH, Suh SH, Baik SJ, Lee HS, Kim J, Lee KY. Characteristics of Intracranial Aneurysms According to Levels of Coronary Artery Calcium. Stroke 2019;50:1403-8.
23. Yao XY, Jiang CQ, Jia GL, Chen G. Diabetes mellitus and the risk of aneurysmal subarachnoid haemorrhage: A systematic review and meta-analysis of current evidence. J Int Med Res 2016;44:1141-55.
24. Nahed BV, DiLuna ML, Morgan T, Ocal E, Hawkins AA, Ozduman K, Kahle KT, Chamberlain A, Amar AP, Gunel M. Hypertension, age, and location predict rupture of small intracranial aneurysms. Neurosurgery 2005;57:676-83; discussion 676-83.
25. Morita A, Kirino T, Hashi K, Aoki N, Fukuhara S, Hashimoto N, Nakayama T, Sakai M, Teramoto A, Tominari S, Yoshimoto T. The natural course of unruptured cerebral aneurysms in a Japanese cohort. N Engl J Med 2012;366:2474-82.
26. Murayama Y, Takao H, Ishibashi T, Saguchi T, Ebara M, Yuki I, Arakawa H, Irie K, Urashima M, Molyneux AJ. Risk Analysis of Unruptured Intracranial Aneurysms: Prospective 10-Year Cohort Study. Stroke 2016;47:365-71.
27. Zhao L, Zhang L, Zhang X, Li Z, Tian L, Wang YX. An analysis of 1256 cases of sporadic ruptured cerebral aneurysm in a single Chinese institution. PLoS One 2014;9:e85668.
28. Hoffman H, Toshkezi G, Swarnkar A, Gould G, Chin LS, Krishnamurthy S. A retrospective comparison of sac and lobe morphology between ruptured and unruptured intracranial aneurysms. J Clin Neurosci 2019;60:88-92.
29. Abboud T, Rustom J, Bester M, Czorlich P, Vittorazzi E, Pinnschmidt HO, Westphal M, Regelsberger J. Morphology of Ruptured and Unruptured Intracranial Aneurysms. World Neurosurg 2017;99:610-7.
30. Zhang Y, Jing L, Liu J, Li C, Fan J, Wang S, Li H, Yang X. Clinical, morphological, and hemodynamic independent characteristic factors for rupture of posterior communicating artery aneurysms. J Neurointerv Surg 2016;8:808-12.
31. Miyata T, Kataoka H, Shimizu K, Okada A, Yagi T, Imamura H, Koyanagi M, Ishibashi R, Goto M, Sakai N, Hatano T, Chin M, Iwasaki K, Miyamoto S. Predicting the growth of middle cerebral artery bifurcation aneurysms using differences in the bifurcation angle and inflow coefficient. J Neurosurg 2023;138:1357-65.
32. Merritt WC, Berns HF, Ducruet AF, Becker TA. Definitions of intracranial aneurysm size and morphology: A call for standardization. Surg Neurol Int 2021;12:506.
33. Matsukawa H, Kamiyama H, Kinoshita Y, Saito N, Hatano Y, Miyazaki T, Ota N, Noda K, Shonai T, Takahashi O, Tokuda S, Tanikawa R. Morphological parameters as factors of 12-month neurological worsening in surgical treatment of patients with unruptured saccular intracranial aneurysms: importance of size ratio. J Neurosurg 2019;131:852-8.
34. Duan Z, Li Y, Guan S, Ma C, Han Y, Ren X, Wei L, Li W, Lou J, Yang Z. Morphological parameters and anatomical locations associated with rupture status of small intracranial aneurysms. Sci Rep 2018;8:6440.
35. Wang GX, Liu J, Chen YQ, Wen L, Yang MG, Gong MF, Zhang D. Morphological characteristics associated with the rupture risk of mirror posterior communicating artery aneurysms. J Neurointerv Surg 2018;10:995-8.
36. Jiang H, Shen J, Weng YX, Pan JW, Yu JB, Wan ZA, Zhan R. Morphology Parameters for Mirror Posterior Communicating Artery Aneurysm Rupture Risk Assessment. Neurol Med Chir (Tokyo) 2015;55:498-504.
37. Xu WD, Chen RD, Hu SQ, Hou YY, Yu JS. Morphological evaluation of the risk of posterior communicating artery aneurysm rupture: a mirror aneurysm model. J Neurosurg 2023;138:185-90.
38. Varble N, Tutino VM, Yu J, Sonig A, Siddiqui AH, Davies JM, Meng H. Shared and Distinct Rupture Discriminants of Small and Large Intracranial Aneurysms. Stroke 2018;49:856-64.
39. Juvela S, Poussa K, Lehto H, Porras M. Natural history of unruptured intracranial aneurysms: a long-term follow-up study. Stroke 2013;44:2414-21.
40. Backes D, Rinkel GJ, Laban KG, Algra A, Vergouwen MD. Patient- and Aneurysm-Specific Risk Factors for Intracranial Aneurysm Growth: A Systematic Review and Meta-Analysis. Stroke 2016;47:951-7.