Center for In-Vivo

Hyperpolarized Gas MR Imaging

CP421, Polarized Gas Targets and Polarized Beams: Seventh International Workshop
edited by Roy J. Holt and Michael A. Miller
©1998 The American Institute of Physics 1/56396-700-6/98/$15.00

Polarized Noble Gas MRI

James R. Brookeman^1,2, John P. Mugler III^1,2,
Paul Bogorad⁵, Thomas M. Daniel³, Eduard E. de Lange¹,
Bastiaan Driehuys⁵, Jack Knight-Scott¹, Therese Maier¹, Jonathon D.
Truwit⁴, Gordon Cates⁵, and William Happer⁵

Departments of Radiology¹, Biomedical Engineering², Surgery³
and Internal Medicine⁴, University of Virginia Health Sciences Center,
Charlottesville, Virginia 22908, and Department of Physics⁵, Princeton
University, Princeton, New Jersey 08544

Abstract. The development of convenient methods to polarize liter quantities of the noble gases helium-3 and xenon-129 has provided the opportunity for a new MRI method to visualize the internal air spaces of the human lung. These spaces are usuall y poorly seen with hydrogen-based MRI, because of the limited water content of the lung and the low thermal polarization of the water protons achieved in conventional magnets. In addition, xenon, which has a relatively high solubility and a sufficiently p ersistent polarization level in blood and biological tissue, offers the prospect of providing perfusion images of the lung, brain and other organs.

Techniques employed for creating laser-polarized spin targets for high-energy physics have been adapted (1) to polarizing liter quantities of the noble gases ³He and ¹²⁹Xe, that when inhaled make possible magnetic resonance (MR) imag es of the human lung spaces. Although the noble gas density achieved in the lung is considerably less than the water density in most human tissues, the gain in magnetization achieved by the laser polarization process is so great (>10⁵) that ex cellent quality lung space images can be obtained with ¹²⁹Xe and ³He MRI.

MR STUDIES WITH ¹²⁹XE

Early MRI studies of rats and mice with ¹²⁹Xe (2) showed the potential for polarized noble gas MRI and provided the incentive to develop laser gas polarizers suitable for human studies. In April 1996 a prototype laser diode-array polarizer for ¹²⁹Xe was assembled by the Princeton physics group at the University of Virginia. Enriched xenon gas (71% ¹²⁹Xe), polarized via spin exchange with an optically pumped rubidium vapor, was accumulated as a solid in a cold finger from a gas stream of 1% Xe, 1% N₂, and 98% ⁴He. The frozen xenon was then sublimed and collected in a 500 cm³ plastic bag for delivery to the subject positioned in a 1.5 Tesla whole-body MR

scanner (Magnetom Vision, Siemens Medical Systems, Iselin, NJ). With polarization levels of approximately 2% cross-sectional images of the lung gas-spaces were acquired during a 12 second breathhold (see Figure 1). In separate MR spectroscopy experiments , three dissolved-phase peaks were detected from the chest (see Figure 2), and a single prominent peak, shifted 196 ppm from the gas peak, was observed in the human brain (3).

MR STUDIES WITH ³HE

In April 1997 lung images with ³He were obtained at the University of Virginia (see Figure 3), employing an investigational noble gas polarizer (Magnetic Imaging Technologies Inc., Durham, NC). This system, which is similar in operation to th e ¹²⁹Xe polarizer, is capable of producing 2-3 liters of 15-25% polarized ³He gas in a

3-4 hour accumulation. In a typical ³He MR lung study, the volunteer inhales a liter of ³He from a small plastic bag, and 10 to 15 contiguous 1-cm sections of the lung field are obtained during a 15-22 second breathhold. The first human lung studies with optically polarized ³He reported in 1 996 (4,5,6) showed healthy volunteers with a near homogeneous MR signal observed throughout the lung field, except for areas where the gas was displaced by normal structures, such as vessels. In patient studies (5), lesions were apparent as MR signal de fects, and obstructive lung disease was depicted with severely inhomogeneous signal intensity. Our study of a patient with chronic obstructive pulmonary disease (COPD) has similar findings (see Figure 4) with significant signal voids that correspond to ventilation defects seen in a nuclear medicine study of the same patient.

SUMMARY

The precise role that this type of imaging procedure might play in a patient's evaluation and the relative merits of ³He versus ¹²⁹Xe are yet to be determined; but the initial results are sufficiently encouraging to warrant further s tudy. The possibility of imaging the dissolved phases of ³Xe in the lung and brain to provide images of tissue perfusion is particularly attractive. This would require higher polarization levels (> 20%) and probably some form of echo-train pu lse sequence to take advantage of the long T₂ values of these dissolved phases.

Continued improvements in the technology of laser polarized noble gas MRI can be expected to enhance the potential diagnostic value of this new method of imaging, which is readily transferable to many of the MRI centers located worldwide with simple addit ions to the basic MR platform. It is also possible that very low field MR systems could be developed to reduce the cost of the procedure, since the MR signal strength for this technology does not depend primarily on the value of the applied magnetic fiel d. Furthermore, for some applications, it may be advantageous to introduce polarized dissolved-phase xenon directly into blood or tissue using a compatible solvent carrier as proposed by Bifone et al. (7). This could considerably reduce the burden of xe non in the lung, and also significantly increase the MR signal for regional perfusion studies.

ACKNOWLEDGMENTS

The authors thank Joe Camaratta, Wilhelm Dürr, Andreas Potthast, and Phillip Belcher of Siemens Medical Systems for important contributions in bringing the broadband imaging/spectroscopy option and the helium and xenon RF coils into operation, and Stuart Berr, John Christopher, Denise Hinton, and Shella Keilholz, for valuable assistance with the MR studies. This research was supported by the University of Virginia Pratt Fund, the Dean of the Medical School, Dr. Robert M. Carey, and Siemens Medical System s.

REFERENCES

1. Anthony P.L. et al., Phys. Rev. Lett. 71, 959-962 (1993), Happer W., Miron E., Schaefer S., Schrieber D., van Wijngaarden W., Zeng X., Phys. Rev. A. 29, 3092-3110 (1984), Driehuys B, Cates G., Miron E., Sauer K., Walter D., and Happer W., Appl. Phys. Lett. 69, 1668-1670 (1996).

2. Albert M., Cates G., Driehuys B., Happer W., Saam B., Springer Jr, C., Wishnia A., Nature 370, 199-201 (1994), Sakai K., Bilek A., Oteiza E., Walsworth R., Balamore D., Jolesz F., and Albert M., J. Magn. Reson. Series B. 111 , 300-304 (1996), Wagshul M, Button T., Li H., Liang Z., Springer C., Zhong K., Wishnia A., Magn. Reson. Med. 36, 183-191 (1996).

3. Mugler III, J., Driehuys B., Brookeman J., Cates G., Berr S., Bryant R., Daniel T., de Lange E., Downs III, J. , Erickson C., Happer W., Hinton D., Kassel N., Maier T., Phillips C., Saam B.T., Sauer K., Wagshul M., Mag. Reson. Med. 37, 809-815 (1997).

4. MacFall, J., Charles H., Black R., Middleton H., Swartz J., Saam B., Driehuys B., Erickson C., Happer W., Cates G., Johnson G., Ravin C., Radiology, 200, 553-8 (1996).

5. Bachert P., Schad L., Bock M., Knopp M., Ebert M., GroBman T., Heil W., Hofmann D., Surkau R., and Otten E., Magn. Reson. Med. 36, 192-196 (1996).

6. Kauczor H., Hofmann D., Krietner K., Nilgens H., Surkau R., Heil W., Potthast A., Knopp M., Otten E., and Thelen M., Radiology 201, 564-8 (1996).

7. Bifone A, Song Y-Q, Seydoux R, Taylor R, Goodson B, PietraB T, Budinger T, Navon G, Pines A. Proc Natl Acad Sci 93, 12932-12936 (1996).