E-005 Non-invasive intracranial pressure estimation in idiopathic intracranial hypertension using 4D flow MRI and computational fluid dynamics

• Published in: Journal of neurointerventional surgery Vol. 16; no. Suppl 1; p. A83
• Main Authors: Kazemi, A, Negahdar, M, Amini, A, Abecassis, I
• Format: Journal Article
• Language: English
• Published: BMA House, Tavistock Square, London, WC1H 9JR BMJ Publishing Group Ltd 01.07.2024; BMJ Publishing Group LTD
• Subjects: Boundary conditions; Fluid dynamics; Hypertension; Magnetic resonance imaging; SNIS 21st annual meeting electronic poster abstracts; Software
• ISSN: 1759-8478; 1759-8486
• DOI: 10.1136/jnis-2024-SNIS.110

IntroductionIdiopathic intracranial hypertension (IIH) is characterized by elevated intracranial pressures (ICP), leading to severe morbidity, headaches and blindness. This condition often coincides with cerebral venous sinus stenosis, particularly at the transverse-sigmoid junction, causing venous outflow obstruction and elevated ICPs. While venous sinus stenting has emerged as a potential intervention, it necessitates invasive procedures like cerebral angiography and venous manometry to evaluate venous pressure gradients, posing significant risks. Four-dimensional flow magnetic resonance imaging (4D Flow MRI) offers a non-invasive method to assess the underlying hemodynamics in IIH. Additionally, Computational Fluid Dynamics (CFD) can be used to simulate patient-specific blood flow dynamics.MethodsWe performed a series of experiments to analyze data from time-of-flight (TOF) and 4DFlow MRI scans, as well as CFD in Ansys Workbench for 2 patients with IIH and 2 control subjects. TOF images were overlaid onto the 4D flow MRI scans to ensure a comprehensive understanding of the field of view and anatomical details. The GTFlow software was utilized to draw contours around the region of interest and save them as contour masks. These masks were subsequently applied to the 4D flow MRI to isolate and extract the desired region. We carried out level set segmentation to obtain anatomical details, accurately separating critical structures from surrounding anatomy using the VTK algorithm. The resulting data pieces were saved as a stereolithography (STL) file.ResultsWe created a high-resolution mesh (figure 1a), and the surfaces for boundary conditions were defined using the Ansys software. The mesh served as the foundation for the CFD model, facilitating the solution of fluid dynamics equations. 4D flow MRI data provided velocity profile and flow waveform (figure 1b), which were used as boundary conditions in CFD simulation. Following the simulation, the CFD results were compared with pressures from the Bernoulli equation, using 4D flow MRI velocities (figure 1d) in TS-SS section. The ICP signals obtained from real-time Philips monitor in TS-SS section is shown in (figure 1c). CFD pressure contours at peak systolic time points are depicted in (figure 1e). The heart rate for the patient was 150 (bpm) and the error between CFD predictions and Bernoulli pressure estimation was approximately 27%.ConclusionOur study demonstrates the potential of integrating 4D Flow MRI and CFD for non-invasive ICP in IIH, offering a useful alternative to current gold-standard, invasive diagnostics. This approach mitigates patient risk and provides details on the hemodynamic parameters underlying IIH pathophysiology.Abstract E-005 Figure 1Disclosures A. Kazemi: None. M. Negahdar: None. A. Amini: None. I. Abecassis: None.