A theoretical framework to model DSC-MRI data acquired in the presence of contrast agent extravasation.

Dynamic susceptibility contrast (DSC) MRI methods rely on compartmentalization of the contrast agent such that a susceptibility gradient can be induced between the contrast-containing compartment and adjacent spaces, such as between intravascular and extravascular spaces. When there is a disruption of the blood-brain barrier, as is frequently the case with brain tumors, a contrast agent leaks out of the vasculature, resulting in additional T(1), T(2) and T*(2) relaxation effects in the extravascular space, thereby affecting the signal intensity time course and reducing the reliability of the computed hemodynamic parameters. In this study, a theoretical model describing these dynamic intra- and extravascular T(1), T(2) and T*(2) relaxation interactions is proposed. The applicability of using the proposed model to investigate the influence of relevant MRI pulse sequences (e.g. echo time, flip angle), and physical (e.g. susceptibility calibration factors, pre-contrast relaxation rates) and physiological parameters (e.g. permeability, blood flow, compartmental volume fractions) on DSC-MRI signal time curves is demonstrated. Such a model could yield important insights into the biophysical basis of contrast-agent-extravasation-induced effects on measured DSC-MRI signals and provide a means to investigate pulse sequence optimization and appropriate data analysis methods for the extraction of physiologically relevant imaging metrics.