(522h) Comparative Computational Fluid Dynamics Analysis of Pulsatile and Steady State Flow in Ventricular Catheters

Hydrocephalus treatment involves using a ventricular catheter (VC) placed inside the lateral ventricles in the brain to remove excess cerebrospinal fluid (CSF); however, historically, VCs have been fraught with high mechanical failure rates. The current failure rate of ventricular catheters is 85% within ten years^1,2 due predominantly to mechanical obstructions of the drainage holes. These obstructions are caused by tissue obstruction and cellular adhesion. Prior work has hypothesized that CSF flow into the VCs causes cells to flow into the drainage holes, resulting in transient blockage and eventual failure. Previous research^3-6 has utilized computational fluid dynamic simulations, which applied steady-state modeling to the ventricular system to analyze CSF vector flow patterns into VC models. The ventricular system, however, produces pulsatile CSF due to oscillatory waveforms from arterial blood flow. The relationship between oscillatory^7,8 CSF in a hydrocephalic ventricular system and the VC is yet to be examined in the literature.

Deidentified, patient-specific MRI and CT scans were used from our collaborative medical centers at Cincinnati Children's Hospital Center and Children's Hospital of Alabama to create 3D hydrocephalic ventricular renders for CFD simulation. Ventricular catheters were inserted into the lateral ventricles using anterior trajectories. Physiological boundary conditions such as CSF secretion, heart rate, respiration, and intracranial pressure were assigned to the ventricular domain with a constant pressure outlet at the VC exit. A laminar steady-state flow model was compared to a transient model to analyze fluid behavior inside the drainage holes and lumen. Comparative analysis was performed using Poly-Hexcore, Tetrahedral, and Polyhedral element meshing to determine which volume element resulted in steady convergence and accurate gradient approximation. Uniform boundary layers were applied at all VC surface walls, and a specific growth rate for each element type was combined for mesh independence testing. Body-specific meshing was implemented to ensure a fine mesh transition between the free-stream ventricular domains and near-wall VC surfaces. Flow parameters, including mass flow rate, shear stress contours, pressure gradients, and velocity streamlines, were quantified in the catheter drainage holes and lumen.

The data show that transient pulsatile flow resulted in higher mass flow rates and shear stresses in the drainage holes and lumen compared to steady state flow during peak diastole, with a steady decrease in flow rates during systole for highly enlarged ventricles. However, for smaller ventricular morphologies, there was a decrease in mass flow in the holes, with much of the flow being redirected into other areas of the ventricular compartment due to highly complex morphology. In some cases, steady-state values predicted similar flow gradients for smaller ventricles that were touching the VC surface when compared to transient flow during the cardiac cycle. This observation was not seen in ventricles of larger size and morphology.

In certain complex cases such as loculated hydrocephalus, reverse flow boundary layer separation was observed within certain drainage holes of the catheter, attributed to interaction with specific regions of the ventricular body and choroid plexus. This phenomenon was not uniform across all catheter or anatomical interfaces, presumably due to variations in their spatial relationships. By incorporating patient-specific ventricular morphology and physiological boundary conditions, we aim to analyze how CSF dynamics are transiently affected over the course of the VC's life cycle inside the brain. Future work will analyze how ventricular elasticity⁹ plays a role in waveform degradation around the VC during hydrocephalus.

1. Harris CA, McAllister JP 2nd. What we should know about the cellular and tissue response causing catheter obstruction in the treatment of hydrocephalus. Neurosurgery. 2012 Jun;70(6):1589-601
2. Harris CA, McAllister JP 2nd. What we should know about the cellular and tissue response causing catheter obstruction in the treatment of hydrocephalus. Neurosurgery. 2012 Jun;70(6):1589-601; discussion 1601-2.
3. Galarza M, Giménez Á, Valero J, Pellicer OP, Amigó JM. Computational fluid dynamics of ventricular catheters used for the treatment of hydrocephalus: a 3D analysis. Childs Nerv Syst. 2014 Jan;30(1):105-16
4. Galarza M, Giménez Á, Valero J, Pellicer O, Martínez-Lage JF, Amigó JM. Basic cerebrospinal fluid flow patterns in ventricular catheters prototypes. Childs Nerv Syst. 2015 Jun;31(6):873-84
5. Lin J, Morris M, Olivero W, Boop F, Sanford RA. Computational and experimental study of proximal flow in ventricular catheters. Technical note. J Neurosurg. 2003 Aug;99(2):426-31
6. Galarza M, Giménez Á, Pellicer O, Valero J, Amigó JM. New designs of ventricular catheters for hydrocephalus by 3-D computational fluid dynamics. Childs Nerv Syst. 2015 Jan;31(1):37-48
7. Gholampour, S., Fatouraee, N. Boundary conditions investigation to improve computer simulation of cerebrospinal fluid dynamics in hydrocephalus patients. Commun Biol 4, 394 (2021)
8. Fillingham P, Rane Levendovszky S, Andre J, Parsey C, Bindschadler M, Friedman S, Kurt M, Aliseda A, Levitt MR. Patient-specific computational fluid dynamic simulation of cerebrospinal fluid flow in the intracranial space. Brain Res. 2022 Sep 1;1790:147962.
9. Solamen LM, McGarry MDJ, Fried J, Weaver JB, Lollis SS, Paulsen KD. Poroelastic Mechanical Properties of the Brain Tissue of Normal Pressure Hydrocephalus Patients During Lumbar Drain Treatment Using Intrinsic Actuation MR Elastography. Acad Radiol. 2021 Apr;28(4):457-466