Abstract
Cerebrospinal fluid (CSF) plays a vital role in the immunological support, structural protection and metabolic homeostasis of the central nervous system (CNS). CSF is a clear fluid that surrounds the entire brain and spinal cord and has a viscosity approximately that of water. It moves in an oscillatory manner, with zero net flow, in synchrony with intracranial blood pulsations. A detailed understanding of CSF dynamics may improve treatment of several CNS diseases and help to optimize CSF system-based CNS therapeutics. The importance of CSF dynamics has been investigated in several CNS diseases including syringomyelia [1], Alzheimer's disease [2], Chiari malformation [3], and hydrocephalus [4]. Recent studies have examined the possible role of CSF as a conduit for distribution of therapeutic molecules to neuronal and glial cells of CNS tissues [5]. There is a need to develop a CSF hydrodynamic simulator with a realistic geometry and CSF flow distribution. Such a simulator will allow testing and optimization of delivery regimens for CNS therapeutics in development. Since these therapies often require testing on non-human primates (NHPs), our approach was to develop a subject-specific numerical model of CSF hydrodynamics in a cynomolgus monkey, a commonly used species for these studies. The focus of this numerical model was accurate representation of the spinal subarachnoid space (SAS), CSF flow rate, and waveform distribution as intrathecal infusion is primarily conducted within the spine. METHODS A healthy four-year-old adult male cynomolgus monkey with a weight of 4.39 kg was selected for the study. Total MRI imaging time was 81 minutes. High-resolution T2-weighted anatomic MR images (375 μm isotopic) were collected on a Philips 3T scanner. Phase-contrast MRI (PCMRI) measurements were collected with retrospective ECG gating and 24 heart phases were reconstructed over the cardiac cycle. Slice location for each scan was oriented perpendicular to the CSF flow direction with slice planes intersecting vertebral discs. These locations included the foramen magnum (FM) and vertebral disks located between the C2-C3, T4-T5, T10-T11 and L2-L3 vertebral levels (Figure 1a). CSF flow was quantified for each of the axial locations shown in Figure 1b. The T2-weighted images were semi-automatically segmented using ITK-snap software. The commercial finite volume computational fluid dynamic (CFD) solver ANSYS FLUENT V17.2 was used to solve the continuity and Navier-Stokes equations. To reproduce the non-uniform distribution of CSF flow along the spine (Figure 1b), a non-uniform deformation of the computational mesh was implemented at each time step. To verify our numerical results, independence studies were carried out to determine the effect of cycle, mesh size and time-step size on velocity results. Results were analyzed based on the second cycle with 11M cells and a 10 ms time-step size.