INTRODUCTION

Magnetic resonance imaging (MRI) of pulmonary arteries has a significant role in the understanding of pulmonary and cardiac diseases. However, because of vessel size and respiratory motion, pulmonary angiography in small rodents is still challenging and requires high spatial and temporal resolutions. The reference method is micro CT scanner. However, this method is not suitable for longitudinal evaluation of pathologies. Pulmonary MR angiography (MRA) in living mice would therefore provide an important surrogate for cardiovascular research. This study aims at reconstructing MR angiography of mouse lungs with high spatial resolution, due to the use of a self-gated UTE 3D sequence with a pseudo-random encoding [1] combined with the injection of iron nanoparticles.

METHODS

MRI experiments were performed on a 7T Bruker system (Germany), using a 4-phased array coil in a volume configuration (20mm diameter). C57BL/6 mice were anesthetized with isoflurane. Before mice positioning in the magnet, 100µmol Fe/kg of ultrasmall super paramagnetic iron oxide (USPIO, Ferumoxytol) was injected through the tail vein. A 3D self-gated UTE sequence was used : TR/TE=3.5/0.081 ms, Flip angle=15°, 5 data points sampled for self-gated signal (SG), 600 000 projections (30000 x 20 repetitions) distributing according to the pseudo Golden angle method, TA=35min. A field of view of 16x16x16 mm was used and a matrix size of 128x128x128 mm resulting in an isotropic resolution of 125 µm. From the SG data, the respiratory signal was extracted and 30 images were reconstructed along the respiratory cycle. Data during the respiratory movement were either not used to prevent motion artifact, and to improve spatial resolution (cases « 15 images » and « 1 image » in Figure 1), or used to show the impact of the respiratory movement on angiography images (case « 30 images » in Figure 1).

RESULTS

When all the images acquire during the respiratory cycle are used, the signal increases but the spatial resolution decreases due to motion. Consequently, a compromise has to be made between the quantity of data from the respiratory peak used for reconstruction and the signal to noise ratio. Figure 2 shows the images obtained with three different mount of images, as described previously (30, 15 or 1 images). Image A illustrates that including the data acquired during the respiratory peak generates blurriness due to motion, compared to the one reconstructed with only the 15 images centered during the stable phase (B). In that latter case, a highly-resolved MR angiography was obtained, enabling the detection of small pulmonary vessels. Finally, using only one image of the respiratory cycle resulted in noise. Therefore, pulmonary angiographies were reconstructed using the 15 images centered during the stable phase. The image quality of lung parenchyma in our study is consistent (see figure 3), in which small blood vessels and their branches can be detected (red arrows), as well as bronchi (blue arrows).

DISCUSSION

The present work shows the efficiency of using a 3D UTE SG sequence coupled with iron nanoparticles injection, to reconstruct highly-resolved pulmonary angiography of mice. By only taking data acquired during the stable phase of the respiratory cycle, image quality is improved. In 35 min, high quality angiography of the lungs, comparable to micro CT scanner, can be reconstructed, allowing small structure detection like vessels and bronchi.

CONCLUSION

Here, we demonstrated the feasibility of providing highly-resolved free-breathing pulmonary angiography, with an isotropic spatial resolution of 125 µm in mice.

Acknowledgements

No acknowledgement found.