Geometrical and Hemodynamic Characteristic Changes of Small Anterior Communicating Artery Aneurysms during Follow-ups in a Retrospective Analysis
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.329617
Source of Support: None, Conflict of Interest: None
Keywords: Anterior communicating artery, computational fluid dynamics, geometry, intracranial aneurysms, rupture
Unruptured intracranial aneurysm is a common cerebrovascular disease with a prevalence of 3–5% in the general population. An incidental finding of asymptomatic aneurysms poses a great challenge to decision-making due to the lack of reliable methods for accurately assessing the risk of aneurysm rupture or the intervention itself.,,,, In the international unruptured intracranial aneurysm study, small aneurysms (less than 7 mm) were reported to have a very low risk of rupture, the overall risk of rupture was 0.7% per year, and the 5-year cumulative rupture rate was 0% among patients with anterior circulation aneurysms without previous subarachnoid hemorrhage. The average annual rate of rupture of small unruptured aneurysms is 0.54%, however, small anterior communicating artery (ACoA) aneurysms have a greater risk of rupture than small aneurysms of other locations. About one-half of ruptured ACoA aneurysms are small (less than 5 mm) in the clinical practice and ruptured ACoA aneurysms more often have smaller sizes.,,
Several studies have reported a variety of geometrical and hemodynamic metrics associated with the rupture of intracranial aneurysms in cross-sectional studies.,,,,,,,,,,, However, geometrical and hemodynamic characteristics of small unruptured ACoA aneurysms during the observations are rarely reported., In this study, we retrospectively collected patients with small unruptured ACoA aneurysms that were not repaired, who had serial follow-ups. We performed a retrospective analysis to investigate the geometrical and hemodynamic changes of small ACoA aneurysms during follow-ups.
The study was approved by the IEC was approved by which Institute in November, 2019, and patient informed consent was waived due to the retrospective nature of the study. We searched clinical data on unruptured ACoA aneurysms from the Electronic Medical Record System from January 2016 to March 2019 in four tertiary hospitals. Patients were included as per the following criteria: patients with unruptured aneurysms without repair at initial diagnosis, the maximal diameter of the aneurysm being less than 5 mm, the aneurysms that are located at the ACoA, and patients who had imaging follow-ups for more than 12 months. Patients with multiple aneurysms or any other ruptured aneurysms were excluded. Most patients with unruptured ACoA aneurysms were treated with endovascular coiling or surgical clipping because unruptured ACoA aneurysms have a higher risk rupture than unruptured aneurysms of other locations and are recommend to be treated to prevent the rupture., It is controversial to treat small (<5 mm) unruptured aneurysms because the natural history of small unruptured aneurysms remains unclear. Five patients with small unruptured ACoA aneurysms that were not repaired were identified. In one patient, the magnetic resonance angiography (MRA) image detected unruptured ACoA aneurysm, however, it was regarded as a normal finding by radiologists. This patient was not diagnosed with an ACoA aneurysm until subarachnoid hemorrhage occurred. Four patients were concerned with treatment risk and received clinical observation.
The unruptured ACoA aneurysm of patient 1# (54-year-old male, with history of smoking) was diagnosed on December 21, 2015, and the first, second, and third follow-ups were made at 7 months, 16 months, and 23 months, respectively after the initial diagnosis. The aneurysm ruptured after 2 days of the last imaging examination. The unruptured ACoA aneurysm of patient 2# (70-year-old female) was detected by MRA on April 6, 2017, and the second follow-up was done 15 months after the initial diagnosis. The third imaging examination was performed after the aneurysm rupture. The unruptured ACoA aneurysm of patient 3# (59-year-old male, with history of hypertension) was diagnosed on May 29, 2015, and the first, second, and third follow-ups were done at 16 months, 27 months, and 33 months, respectively after the diagnosis. The unruptured ACoA aneurysm of patient 4# (67-year-old female) was diagnosed on April 25, 2016, and the follow-up was 29 months after the diagnosis. The unruptured ACoA aneurysm of patient 5# (70-year-old female) was diagnosed on March 22, 2017, and the follow-up was 14 months after the initial diagnosis.
Computed tomography angiography (CTA) or MRA images at each follow-up were collected. CTA was performed using a 64-row multidetector CT scanner (Philips Healthcare, The Netherlands) or a 3T MR scanner (Philips Healthcare, The Netherlands) with standard technique (slice thickness: 0.5–0.7 mm, in-plane spatial resolution: 512 × 512 pixels). CTA or MRA data of each patient was subsequently read into Mimics 15 (Materialise, Belgium) to perform image segmentation and construct a 3-D geometrical model of the aneurysm along with the cerebral arterial network.
Computational fluid dynamic modeling
Aneurysmal region of interest was isolated, containing the aneurysm sac, parent artery, and major adjacent arteries. The isolated geometrical model was subsequently imported into ANSYS ICEM CFD 16.0 to generate a high-resolution mesh model. Blood flow was assumed to be an incompressible fluid with a density of 1060 kg/m3 governed by the unsteady mass conservation and Navier-Stokes (N-S) equations. The classical Carreau model was employed to dynamically relate blood viscosity to shear rate to account for the non-Newtonian rheology of blood. All models were assumed to have rigid walls to which the nonslip flow conditions were imposed. Population-averaged flow velocity waveforms simulated by a one-dimensional model of the cerebral arterial network was used to prescribe the inflow boundary conditions for each aneurysm model. The outlets were supported by resistance boundary conditions which were adjusted to yield physiological flow division among the efferent arteries. The mass conservation and N-S equations were discretized and solved in ANSYS CFX 16 where an element-based finite volume method was employed. Second-order schemes were adopted for both spatial discretization and time integration. The numerical time step was fixed at 0.001 s. Each set of numerical simulation lasted for five cardiac cycles to let the numerical solution converge to a periodic one, with the results obtained in the last cardiac cycle being analyzed and reported.
Geometrical and hemodynamic data measurement
The following morphological parameters were calculated based on the reconstructed geometrical model: parent artery size, aneurysm size (maximum perpendicular distance between the neck plane and aneurysm dome), aspect ratio (the ratio of aneurysm depth to aneurysm neck width), and size ratio (the maximum height of aneurysm sac over the parent vessel diameter). The model-simulated time-varying wall shear stress (WSS) was analyzed over a cardiac cycle to derive the time-averaged wall shear stress (TAWSS) and oscillatory shear index (OSI), which have been found to associate with the development and progression of cerebral aneurysms.,,
Geometrical and hemodynamic parameters during follow-ups
Five small unruptured ACoA aneurysms were included in the study. Aneurysm rupture occurred in two patients. The geometrical and hemodynamic parameters of aneurysms at each follow-up are summarized in [Table 1]. The volume, height, and size ratio of the ruptured aneurysms increased gradually during follow-up, whereas the geometrical parameters showed little changes in the unruptured aneurysms.
Geometrical and hemodynamic characteristics during follow-ups
Geometrical characteristics in two ruptured aneurysms and three stable aneurysms at each follow-up are shown in [Figure 1]. Ruptured aneurysms were stable in the first several months of follow-up, but later experienced evident growth featured mainly by the formation and rapid growth of a bleb in the neck or dome region [Figure 1]a and [Figure 1]b. Unruptured ACoA aneurysms maintained a stable morphology without any growth during the serial follow-ups [Figure 1]c and [Figure 1]e.
Hemodynamic characteristics in two ruptured aneurysms and one stable aneurysm are shown in [Figure 2]. The presence of a strong flow jet impinging on the aneurysm neck or dome was detected at the first follow-up in the two ruptured aneurysms. The later formation and growth of a bleb in the flow-impinging region significantly altered local flow patterns at the second and third follow-ups, leading to focal distributions of abnormal WSS (high WSS or low WSS combined with high OSI in the bleb or the transition region between the bleb and parent aneurysm). In unruptured aneurysms, the distribution of WSS was relatively smooth, with most WSS values in the aneurysmal region being comparable to those in the parent artery.
To our knowledge, the retrospective analysis is the first to investigate the geometrical and hemodynamic changes of small unruptured ACoA aneurysms during serial follow-ups. We found two small unruptured aneurysms that grew significantly before the rupture. The ruptured aneurysms were featured by the formation and rapid growth of a bleb in the neck or dome region during follow-up, as well as the dynamic changes of focal distributions of high WSS, or low WSS combined with high OSI. The observational results suggest that ACoA aneurysms rupture can occur in small size, especially in the growing aneurysms. Abnormal geometrical and hemodynamic changes potentially contribute to the rupture of small ACoA aneurysms.
Although a wealth of data on the history of unruptured intracranial aneurysms is available in the literature, the true natural history remains incompletely understood because case selection bias occurs in almost all studies.,,,,, Small unruptured aneurysms, especially anterior circulation aneurysms without previous subarachnoid hemorrhage (less than 7 mm), have a very low risk of rupture. A prospective study on the natural course of unruptured intracranial aneurysms in a Japanese Cohort (UCAS Japan) has also shown that an aneurysm with less than 7 mm of size is associated with a low risk of rupture. However, it has been suggested that aneurysms located at the ACoA and those with a daughter sac, are more likely to bleed. In a retrospective study of 474 ruptured ACoA aneurysms, 134 (28.3%) aneurysms were very small (less than 3 mm) and 278 (58.6%) aneurysms were small (less than 5 mm). That is, about one-third of ruptured ACoA aneurysms were very small. In our study, two out of five patients with small ACoA aneurysms experienced rupture during the follow-up. The risk of unruptured ACoA aneurysms should be carefully assessed even if their sizes are small.
Our study on ACoA aneurysms supports the finding of previous cross-sectional studies or longitudinal observations of other unruptured intracranial aneurysms (UIAs).,,, Aneurysms with an irregular shape, higher aspect ratio, and high size ratio are prone to rupture.,,, The geometrical parameter changes may indicate aneurysm growth. Abnormal intra-aneurysmal hemodynamic hemodynamics characterized by low WSS and high OSI were frequently observed in ruptured or unstable aneurysms.,,,,,,, The hemodynamic changes may be both the driving factor and the consequence of the growth. Our longitudinal observation enabled us to trace the geometrical changes before rupture and to quantify the accompanying changes in the intra-aneurysmal hemodynamic environment.
The retrospective pilot study verified that the formation and rapid growth of a bleb can precede aneurysm rupture. The bleb tends to form at the aneurysmal wall exposed to flow impingement and high WSS. The growth of bleb may alter local flow patterns, which further promote growth or rupture. The alternations in flow patterns exhibit aneurysm-specific characteristics depending on the location and morphology of the bleb. These findings indicate that monitoring the aneurysm-specific changes in morphology and hemodynamics facilitates a better understanding of mechanisms underlying aneurysm growth and rupture. The formation and rapid growth of a bleb and the enhancement in flow disturbance featured by a focal distribution of abnormal WSS may be important predictors of aneurysm rupture during follow-up regardless of their small sizes.
A major limitation of the study is a retrospective analysis of a small number of patients to compare the geometrical and hemodynamic changes between the growing/ruptured and stable aneurysms. This is a descriptive statistic of the small sample size, and we could not calculate the annual rupture rate in small ACoA aneurysms. Nowadays, it is difficult to involve a large number of patients with ACoA aneurysms in a prospective longitudinal study due to ethical issues. The interaction between blood flow and wall deformation is ignored in our study to simplify hemodynamic simulation and improve computational efficiency. The radial motion of the cerebral arterial wall is very small and that the overestimation of WSS resulting from rigid-wall assumption is mild.
The retrospective analysis showed that small unruptured ACoA aneurysms may grow and then rupture at a short time of follow-up. Growing bleb, low WSS, and high OSI at the dome and/or high WSS at the neck potentially contribute to the rupture of small ACoA aneurysms. These aneurysms should be considered for the treatment regardless of the small size.
Financial support and sponsorship
The study was supported by Shanghai Municipal Education Commission—Gaofeng Clinical Medicine Grant Support (20171914), the National Natural Science Foundation of China (11972231), the SJTU Medical Engineering Cross-cutting Research Foundation (YG2017MS45 and ZH2018ZDA07), the China Postdoctoral Science Foundation (2018M640385), Shanghai Municipal Commission of Health and Family Planning (201740080), Shanghai Science and Technology Project (18411962700), the National Key Research and Development Project (2016YFC1300800), and the Clinical Research Plan of SHDC (16CR3031A and 16CR2045B).
Conflicts of interest
The authors declare that they have no conflicts of interest pertaining to the study.
[Figure 1], [Figure 2]