Dissolved oxygen minimally affects magnetic susceptibility in biologically relevant conditions

Purpose To investigate the potential of quantitative susceptibility mapping (QSM) with MRI as a biomarker for tissue oxygenation in fat-water mixture. Oxygen molecules (O2) are paramagnetic. This suggests that dissolved O2 in tissue should affect the measured magnetic susceptibility. However, direct measurements of dissolved O2 in tissues is challenging with QSM as the induced change in susceptibility is below the sensitivity of existing algorithms. QSM in regions that contain fat could be sensitive enough to be used as a marker of tissue oxygenation as oxygen has a larger solubility in fat than in water. Methods The relationship between dissolved O2 concentration and magnetic susceptibility was investigated based on MRI measurements using phantoms made of fat-water emulsions. Dairy cream was used to approximate fat-containing biological tissues. Phantoms based on dairy cream with 35 % fat were designed with controlled concentrations of dissolved O2. O2 was bubbled into the dairy cream to reach O2 concentrations above the concentration at atmospheric pressure, while nitrogen was bubbled in cream to obtain O2 concentrations below atmospheric pressure. Magnetic susceptibility was expected to increase, becoming more paramagnetic, as O2 concentration was increased. Results Magnetic susceptibility from MRI-based QSM measurements did not reveal a dependence on O2 concentration in fat-water mixture phantoms. The relationship between susceptibility and O2 was weak and inconsistent among the various phantom experiments. Conclusion QSM in fat-water mixture appears to be minimally sensitive to dissolved O2 based on phantom experiments. This suggests that QSM is not likely to be sensitive enough to be proposed as a marker for tissue oxygenation, as the change in magnetic susceptibility induced by the change in dissolved O2 concentration is below the current detection limit, even in the presence of fat. ### Competing Interest Statement The authors have declared no competing interest. * Gd : Gadolinium Hct : Hematocrit IDEAL : Iterative Decomposition of Water and Fat with Echo Asymmetry and Least Squares Estimation LBV : Laplacian Boundary Value MEDI : Morphology Enabled Dipole Inversion MGRE Multi-Echo Gradient Echo OEF : Oxygen Extraction Fraction QBOLD : Quantitative Blood-Oxygen-Level Dependent QG Quality-Guided QSM : Quantitative Susceptibility Mapping RESHARP : Regularized-Enabled Sophisticated Harmonic Artifact Reduction for Phase data ROI : Region Of Interest STEAM : Stimulated Echo Acquisition Mode