A network model of glymphatic flow under different experimentally-motivated parametric scenarios

Abstract

Rapidly growing evidence demonstrates that flow of cerebrospinal fluid (CSF) through perivascular spaces (PVSs) - annular channels surrounding vasculature in the brain - is a critically-important component of neurophysiology. CSF inflow contributes during physiological conditions to clearance of metabolic waste and in pathological situations to edema formation. However, brain-wide imaging methods cannot resolve PVSs, and high-resolution methods cannot access deep tissue or be applied to human subjects, so theoretical models provide essential insight. We model this CSF pathway as a network of hydraulic resistances, built from published parameters. A few parameters have very wide uncertainties, so we focus on the estimated limits of their feasible ranges by analyzing different parametric scenarios. We identify low-resistance PVSs and high-resistance parenchyma (brain tissue) as the scenario that best explains experimental observations. Our results point to the most important parameters that should be measured in future experiments. Extensions of our modeling may help predict stroke severity or lead to neurological disease treatments and drug delivery methods.