Chihebeddine Dahmani

Variability of hemodynamic parameters using the common viscosity assumption in a computational fluid dynamics analysis of intracranial aneurysms

BACKGROUND In most simulations of intracranial aneurysm hemodynamics, blood is assumed to be a Newtonian fluid. However, it is a non-Newtonian fluid, and its viscosity profile differs among individuals. Therefore, the common viscosity assumption may not be valid for all patients. OBJECTIVE This study aims to test the suitability of the common viscosity assumption. METHODS Blood viscosity datasets were obtained from two healthy volunteers. Three simulations were performed for three different-sized aneurysms, two using measured value-based non-Newtonian models and one using a Newtonian model. The parameters proposed to predict an aneurysmal rupture obtained using the non-Newtonian models were compared with those obtained using the Newtonian model. RESULTS The largest difference (25%) in the normalized wall shear stress (NWSS) was observed in the smallest aneurysm. Comparing the difference ratio to the NWSS with the Newtonian model between the two Non-Newtonian models, the difference of the ratio was 17.3%. CONCLUSIONS Irrespective of the aneurysmal size, computational fluid dynamics simulations with either the common Newtonian or non-Newtonian viscosity assumption could lead to values different from those of the patient-specific viscosity model for hemodynamic parameters such as NWSS.

Hemodynamic effects from coil distribution with realistic coil models in an aneurysm

Because of its minimal invasiveness, coil embolization has become a popular way to treat aneurysms. The main problem with this method, however, is the poor understanding of the hemodynamics in the aneurysm after coil embolization. To improve this situation, we used a finite element method and computational fluid dynamics to investigate how hemodynamic parameters depend on the spatial distribution of coils. A basic model of an internal carotid artery aneurysm was created, and six realistic coil models were considered for the coil geometry. The material properties of the coils were based on the commercially available embolic coil Target 360 series. The results showed that the reduction in blood velocity in aneurysms was closely related to coil distribution. In addition, the coil volume in the neck region and the density of coils near the aneurysmal wall were further important factors for reducing the velocity. Considering the coil distribution may help to prevent aneurysmal recanalization.

High-Resolution C-Arm CT and Metal Artifact Reduction Software: A Novel Imaging Modality for Analyzing Aneurysms Treated with Stent-Assisted Coil Embolization.

BACKGROUND AND PURPOSE: Combination of high-resolution C-arm CT and novel metal artifact reduction software may contribute to the assessment of aneurysms treated with stent-assisted coil embolization. This study aimed to evaluate the efficacy of a novel Metal Artifact Reduction prototype software combined with the currently available high spatial-resolution C-arm CT prototype implementation by using an experimental aneurysm model treated with stent-assisted coil embolization. MATERIALS AND METHODS: Eight experimental aneurysms were created in 6 swine. Coil embolization of each aneurysm was performed by using a stent-assisted technique. High-resolution C-arm CT with intra-arterial contrast injection was performed immediately after the treatment. The obtained images were processed with Metal Artifact Reduction. Five neurointerventional specialists reviewed the image quality before and after Metal Artifact Reduction. Observational and quantitative analyses (via image analysis software) were performed. RESULTS: Every aneurysm was successfully created and treated with stent-assisted coil embolization. Before Metal Artifact Reduction, coil loops protruding through the stent lumen were not visualized due to the prominent metal artifacts produced by the coils. These became visible after Metal Artifact Reduction processing. Contrast filling in the residual aneurysm was also visualized after Metal Artifact Reduction in every aneurysm. Both the observational (P < .0001) and quantitative (P < .001) analyses showed significant reduction of the metal artifacts after application of the Metal Artifact Reduction prototype software. CONCLUSIONS: The combination of high-resolution C-arm CT and Metal Artifact Reduction enables differentiation of the coil mass, stent, and contrast material on the same image by significantly reducing the metal artifacts produced by the platinum coils. This novel image technique may improve the assessment of aneurysms treated with stent-assisted coil embolization.

O-001 A Novel Metal Artefact Removal Software for C-arm CT; A Novel Imaging Modality to Analyze Aneurysms Treated with Stent Assisted Coil Embolization

Purpose Stent assisted coil embolization recently became a common treatment strategy for the wide neck aneurysms. Normal digital subtraction angiography (DSA), however, does not allow visualization of the deployed stents. With a combination of novel Metal Artefact Removal software (MAR) and C-arm CT, the coil mass, stent, and contrast material can be differentiated. In this study, experimental aneurysms were treated with stent assisted coil embolization and image analysis was performed using the C-arm CT and MAR. I. V. injection of the contrast to obtain the images was also performed. Methods A total of 12 experimental aneurysms were created in 6 swine. The aneurysms were treated with coil embolization using stent assisted technique. High Resolution (HR) C-arm CT with intra-arterial injection of the contrast dye (20%) was performed immediately after the treatment and obtained images were processed with MAR. Intra venous contrast injection (non-diluted) from femoral vein was also performed in 3 swine. Visualization of the residual aneurysms as well as coil loops protruded into the stent lumen was evaluated. Two patients previously treated with stent assisted coil embolization underwent the same image acquisition protocol, and the obtained images were evaluated. Results Coil loops protruded into the stent lumen, which were not visible before the MAR due to the significant metal artefact, were well visualised after the MAR. Likewise, Contrast filling into the residual aneurism was observed after the MAR in every aneurysm. Intravenous injection of the contrast material reproduced similar image quality as compared to the intra arterial injections. The C-arm CT images of the two patients showed remarkable reduction of the metal artefacts after the MAR processing. Conclusions The combination of the C-arm CT and MAR enables the differentiation of the coil mass, stent and contrast material by markedly reducing the metal artefact produced by the coils. This novel image modality may contribute to the evaluation of aneurysms treated with stent assisted coil embolization. Abstract O-001 Figure 1 Disclosures I. Yuki: 1; C; Siemens Grant. Y. Kambayashi: 1; C; Siemens Grant. A. Ikemura: 1; C; Siemens Grant. I. Kan: 1; C; Siemens Grant. Y. Abe: None. A. Mohamed: 5; C; Siemens Japan KK. C. Dahmani: 5; C; Siemens Japan KK. T. Ishibashi: 1; C; Siemens Grant. I. Kajiwara: 1; C; Siemens Grant. K. Nishimura: 1; C; Siemens Grant. Y. Murayama: 1; C; Siemens Grant.

Time-resolved magnetic resonance angiography (TR-MRA) for the evaluation of post coiling aneurysms; A quantitative analysis of the residual aneurysm using full-width at half-maximum (FWHM) value

Magnetic resonance image (MRI) is now widely used for imaging follow-up for post coiling brain aneurysms. However, the accuracy on the estimation of residual aneurysm, which is crucial for the retreatment planning, remains to be controversial. The purpose of this study is to evaluate a new post-processing technique that provides improved estimation of the residual aneurysm after coil embolization. One hundred aneurysms on 93 patients who underwent coil embolization for brain aneurysm were evaluated using the 1.5 Tesla time-resolved magnetic resonance angiography (TR-MRA) one year after the treatment. To minimize the inter-observer variability caused by the window level adjustment, an automatic post processing protocol using the full-width at half-maximum (FWHM) value was utilized. The result was then compared with that from the conventional cerebral angiography. Of the 97 aneurysms that underwent both TR-MRA and DSA, 23 (23.7%) showed residual neck / dome during follow-up. After window level adjustment, the size of the parent artery in the TR-MRA was consistent with that in the DSA. The reconstructed Volume Rendering images provided clear contours of the residual aneurysms and contributed to the understanding the configuration of residual aneurysm. The largest and the smallest diameter of the residual aneurysms was larger in the TR-MRA than in the DSA (8.05 vs. 7.72 mm, p = 0.0004; 4.99 vs. 4.19 mm, p = 0.007 respectively). The sensitivity, specificity, and positive and negative predictive values of TR-MRA compared to DSA were 100%, 97%, 73%, and 100%, respectively. Using the FWHM value to optimize the window level adjustment, the size of the residual component observed in the TR-MRA was larger compared to that in the DSA whereas the size of neck and the parent artery showed consistency between the two modalities. This image processing technique can be used as an effective screening tool for evaluating residual component in post-coiling brain aneurysms.

A new combined parameter predicts re-treatment for coil-embolized aneurysms: a computational fluid dynamics multivariable analysis study

Purpose Coil embolization is a minimally invasive method used to treat cerebral aneurysms. Although this endovascular treatment has a high success rate, aneurysmal re-treatment due to recanalization remains a major problem of this method. The purpose of this study was to determine a combined parameter that can be useful for predicting aneurysmal re-treatment due to recanalization. Methods Patient-specific geometries were used to retrospectively analyze the blood flow for 26 re-treated and 74 non-retreated aneurysms. Post-operatively aneurysms were evaluated at 12-month follow-up. The hemodynamic differences between the re-treatment and non-retreatment aneurysms were analyzed before and after coil embolization using computation fluid dynamics. Basic fluid characteristics, rates of change, morphological factors of aneurysms and patient-specific clinical information were examined. Multivariable analysis and logistic regression analysis were performed to determine a combined parameter—re-treatment predictor (RP). Results Among examined hemodynamic, morphological, and clinical parameters, slight reduction of blood flow velocity rate in the aneurysm, slight increase of pressure rate at the aneurysmal neck and neck area, and hypertension were the main factors contributing to re-treatment. Notably, hemodynamic parameters between re-treatment and non-retreatment groups before embolization were similar: however, we observed significant differences between the groups in the post-embolization average velocity and the rate of reduction in this velocity in the aneurysmal dome. Conclusions The combined parameter, RP, which takes into consideration hemodynamic, morphological, and clinical parameters, accurately predicts aneurysm re-treatment. Calculation of RP before embolization may be able to predict the aneurysms that will require re-treatment.

Relationships between geometrical parameters and mechanical properties for a helical braided flow diverter stent

BACKGROUND Although flow diversion is a promising procedure for aneurysm treatment, the safety and efficacy of this strategy have not been sufficiently characterized. Both mechanical properties and flow reduction effects are important factors in the design of an optimal stent. OBJECTIVE We aimed to clarify the contributions of strut size and pitch to the mechanical properties (radial stiffness and longitudinal flexibility) and geometric characteristics (porosity and pore density) related to flow reduction effects. METHODS Crimping and bending behaviors of the stents were simulated with the finite element method. The relationships between the mechanical properties and geometric characteristics were investigated by changing the strut size and pitch. RESULTS Within the porosity range of 79-82%, the radial stiffness of the stent was similarly influenced by either the strut size or pitch. However, the longitudinal flexibility tended to be influenced more by strut size than by pitch. CONCLUSIONS Adjusting the strut size rather than the pitch can change the mechanical properties while minimizing the change in porosity or pore density related to flow reduction effects.

Computational fluid dynamics analysis of tandem carotid artery stenoses: Investigation of neurological complications after carotid artery stenting

BACKGROUND Combined extra- and intracranial carotid artery stenoses, particularly involving multiple lesions, show complex hemodynamic properties and represent a therapeutic dilemma. We used computational fluid dynamics (CFD) to investigate whether insufficient cerebral blood flow (CBF) in a 70-year-old man with tandem stenoses was the cause of aphasia and right hemiparesis after carotid artery stenting (CAS) of the extracranial stenosis. METHOD Three-dimensional digital subtraction angiography (3D-DSA) was performed before and after balloon angioplasty and CAS in the patient. The geometrical and rheological conditions of the carotid arteries were determined, and computational meshes were generated from the patient-specific 3D-DSA datasets. CFD analysis was performed, and hemodynamic parameters such as mass flow, pressure, fractional flow reserve, and streamlines were calculated. RESULTS Post-CAS simulations showed that the percentage of internal carotid artery mass flow from common carotid artery mass flow increased from 9% to 14% and CBF improved by only 5%. CONCLUSIONS CFD analysis suggested that the neurological complications were caused by insufficient CBF rather than embolic events, and in tandem carotid stenoses, CAS for an extracranial lesion alone may not always sufficiently increase CBF. CFD enabled the noninvasive quantitative estimation of the effects of CAS of each stenotic segment on carotid flow.