The ability to visualize whole-brain vasculature is important for quantitative in vivo investigation of vascular malfunctions in cerebral small vessel diseases, including cancer, stroke and neurodegeneration. Dual mode MRA acquisition with superparamagnetic iron oxide nanoparticles (SPION) provides a unique opportunity to systematically compare and synergistically combine both longitudinal (R₁) and transverse (ΔR₂ and ΔR₂^*) relaxation-based MRAs. Through Monte Carlo (MC) simulation [1] and MRA experiments in normal and tumor-bearing animals with intravascular SPION, we validate that the multiplied ΔR₂- and ΔR₂^*-MRAs simultaneously improve the sensitivity to intra-cortical penetrating vessels and reduces vessel size overestimation of ΔR₂^*-MRA. Then direct benefits of the UTE-ΔR₂-ΔR₂^* combined MRA were visualized and quantified for normal and tumor-bearing rat brains.

Normal and tumor-bearing Sprague-Dawley (SD) rats were used for the MRI experiments. MR images of the SD rat brain, using UTE3D sequence (UTE MRA), RARE sequence (ΔR₂ MRA) [2] and FLASH sequence (ΔR₂^* MRA), before and after injection of SPION were acquired on 7T MR scanner (Bruker, Germany). SPION was administered at the dose of 120 μmol/kg for UTE MRA, 240 μmol/kg for ΔR₂^* MRA and 360 μmol/kg for ΔR₂ MRA, respectively. The ΔR₂^* and ΔR₂ values were calculated using the following equation,

$$ \triangle R_2^*\ and\ \triangle R_2\ = \frac{1}{TE} ln \left(\frac{S_{pre}}{S_{post}}\right) $$

where TE is the echo time, and S_pre and S_post are the pre- and post-contrast signal intensities with gradient echo for ΔR₂^* and spin echo for ΔR₂.The processing equation for combining all MRAs is described by:

$$ {{UTE}_{surface\ and\ inner\ area}}+\left[\triangle R_2^*\ \times\ \triangle R_2\right]_{inner\ area} $$

The resulting UTE-ΔR₂-ΔR₂^* combined MRA was used for visualizing the vascular structure with minimized susceptibility artifacts and enhanced sensitivity.

Figure 1 describes the behaviors of ΔR₂, ΔR₂^*, and ΔR₂ × ΔR₂^* as the distance from vessel surface using MC simulation. The red, blue, and purple lines correspond to ΔR₂, ΔR₂^*, and ΔR₂ × ΔR₂^*, respectively. For ΔR₂, the maximum amplitude decreases, while the width of decay tends to slightly increase as the vessel size increases. For ΔR₂^*, the maximum amplitude remains at a constant level, but the width becomes significantly broader with growing vessel size, as seen in Figure1A-1, B-1, C-1, and D-1. Standardized ΔR₂, ΔR₂^*, and ΔR₂ × ΔR₂^* values were superimposed together in Figure 1A-2, B-2, C-2, and D-2 for multiple vessel sizes. The maximum amplitudes of ΔR₂ × ΔR₂^* were significantly higher than those from individual ΔR₂ and ΔR₂^* and the initial fast decay of ΔR₂ × ΔR₂^* followed that of ΔR₂, indicating both improvement in vessel sensitivity and minimization of vessel size overestimation from the multiplicative process of ΔR₂ × ΔR₂^*.

To directly compare each MRA, a line profile analysis was applied to brain region in Figure 2. As shown in Figure 2A-2, the overestimation of vessel size from the ΔR₂^*-MRA (blue) is clear compared to that of ΔR₂-MRA (red) for a relatively large vessel. The combined MRA (yellow) increased the vessel/tissue contrast and reduced the vessel size overestimation, in agreement with the simulation results. Smaller cortical penetrating vessels were rarely detected in UTE-(green) or ΔR₂-MRA (red) in regions shown in Figure 2A-3, but were visible for ΔR₂^*-MRA (blue) and combined MRA (yellow), also in agreement with the simulation results for smaller vessels.

Figure 3 shows MRAs of SD rat with C6 tumors.The enhanced contrast of tumor region was represented well in ΔR₂ image as indicated by white arrow. The tumor region from the ΔR₂ and ΔR₂^* MRAs (Figure 3A-1 and Figure 3A-2) in the cortex was broadly revealed, however, it was difficult to distinguish the tumor boundary in the ΔR₂^*-MRA alone due to the lowered ΔR₂^* pre image signal from the T₂^* effect. The relatively low tumor contrast of the cortical tumor region UTE image is shown in Figure 3A-3. The UTE-ΔR₂-ΔR₂^* combined MRA showed increased tumor contrast in cortical regions, as shown in Figure 6A-4.

As shown in Figure 4, CNRs from both normal and tumor regions were measured for each rat (n = 4) in order to compare the distinguishability of tumors from each MRA. The highest CNR tumor region value was obtained from UTE-ΔR₂-ΔR₂^* combined MRA. In addition, the CNR difference was highest between the normal and tumor regions from UTE-ΔR₂-ΔR₂^* combined MRA.