Background: Understanding molecular transport in the brain is critical to care and prevention of neurological disease and injury. A key question is whether transport occurs primarily by diffusion, or also by convection or dispersion. Dynamic contrast-enhanced (DCE) MRI offers a whole-brain view of transport and the potential for quantitative analysis to determine fundamental transport parameters. However, few DCE-MRI studies have utilized this potential, instead reporting parameters with arbitrary units disconnected from fundamental transport processes.
Methods: In this work, DCE-MRI experimental data is combined with subject-specific finite-element models to quantify transport parameters in different anatomical regions across the whole mouse brain. Effective diffusivity ( ), a transport parameter combining all mechanisms of transport, is determined for each region by minimizing the root mean square error between simulations and data. The resulting sets are compared to apparent diffusivity ( ) to draw conclusions about dominant transport mechanisms in each region.
Results: In the perivascular regions of major arteries, was over 10,000 times greater than . In the brain tissue, constituting interstitial space and the perivascular space of smaller blood vessels, was 10-25 times greater than .
Conclusions: The analysis concludes that convection is present throughout the brain. Convection is dominant in the perivascular space of major surface and branching arteries (Pe > 10,000) and significant to large molecules (>1 kDa) in the combined interstitial space and perivascular space of smaller arteries (not resolved by DCE-MRI). Importantly, this work supports periarterial convection along penetrating and smaller arteries.