Measurement of blood-brain barrier permeability using dynamic contrast-enhanced magnetic resonance imaging with reduced scan time.
Publication Type | Academic Article |
Authors | Bae J, Zhang J, Wadghiri Y, Minhas A, Poptani H, Ge Y, Kim S |
Journal | Magn Reson Med |
Volume | 80 |
Issue | 4 |
Pagination | 1686-1696 |
Date Published | 03/05/2018 |
ISSN | 1522-2594 |
Keywords | Blood-Brain Barrier, Capillary Permeability, Image Interpretation, Computer-Assisted, Magnetic Resonance Imaging |
Abstract | PURPOSE: To investigate the feasibility of measuring the subtle disruption of blood-brain barrier (BBB) using DCE-MRI with a scan duration shorter than 10 min. METHODS: The extended Patlak-model (EPM) was introduced to include the effect of plasma flow (Fp ) in the estimation of vascular permeability-surface area product (PS). Numerical simulation studies were carried out to investigate how the reduction in scan time affects the accuracy in estimating contrast kinetic parameters. DCE-MRI studies of the rat brain were conducted with Fisher rats to confirm the results from the simulation. Intracranial F98 glioblastoma models were used to assess areas with different levels of permeability. In the normal brain tissues, the Patlak model (PM) and EPM were compared, whereas the 2-compartment-exchange-model (TCM) and EPM were assessed in the peri-tumor and the tumor regions. RESULTS: The simulation study results demonstrated that scan time reduction could lead to larger bias in PS estimated by PM (>2000%) than by EPM (<47%), especially when Fp is low. When Fp was high as in the gray matter, the bias in PM-PS (>900%) were larger than that in EPM-PS (<42%). The animal study also showed similar results, where the PM parameters were more sensitive to the scan duration than the EPM parameters. It was also demonstrated that, in the peri-tumor region, the EPM parameters showed less change by scan duration than the TCM parameters. CONCLUSION: The results of this study suggest that EPM can be used to measure PS with a scan duration of 10 min or less. |
DOI | 10.1002/mrm.27145 |
PubMed ID | 29508443 |
PubMed Central ID | PMC6340058 |