| Direct Molecular Solution of Hyperpolarised Gases through Hollow Fiber Membranes |
Introduction Xenonizer principles Xenonizer Prototype Comparisson of Dissolving Methods
First Images in a 1.5T clinical Scanner Conclusions and Outlook References
MRI as a diagnostic tool is increasingly employing functional contrast agents to study or contrast entire mechanisms. Hence the production of specialized contrast agents is of great importance.
Conventional MRI contrast agents (e.g. Gd-DTPA) do (and must) not pass membranes in the body due to their large size. Xenon atoms are small enough to pass such barriers (e.g. Blood-brain barrier). Therefore dissolved hyperpolarized (HP) Xenon is an interesting candidate for a new (free diffusive) MRI contrast agent [1]. To dissolve Xenon in blood via inhalation is quite problematic, due to the presence of oxygen and the slow passage into the cardiovascular system [2].
These problems can be overcome by dissolving Xe in a suitable carrier liquid and injection [3].
So far HP-Xe is typically frozen out, vaporized in the presence of the liquid and dissolved by shaking. This process is hard to control and not suitable to be applied in hospitals. We developed a method to dissolve HP-Xe easily and continuously in different liquids using oxygenator membranes modules.
Idea:
Usage of hollow-fiber oxygenator membranes (like in heart-lung machines) in
special modules to dissolve HP-Xenon in various biocompatible solvents.
Principle:
The solvent is circulated through the membrane module from a reservoir driven by
a nonmagnetic membrane pump. At the same time HP-Xenon counter-flows through the
hollow membranes and dissolves into the liquid. When the liquid is enriched with
enough Xenon it can be removed and injected as contrast agent.
To study the dissolution process or to conduct other spectroscopic experiments the reservoir can be put into a NMR probe.
Two prototypes (“Xenonizer”) are already realized (patent pending [4] ).
With this prototype it was shown that Xenon can be dissolved molecularly in blood (or autologous plasma) and in other liquids that could be used as contrast agents. The signals were referenced to the Xenon gas signal from an additional tube inside the probe.
|
|
|
|
|
1st Prototyp |
Spectra for different solvents |
Immediatly signal built up in Lipofundin |
The dissolution process occurs fast, as the maximum signal amplitude is reached already about 10 seconds after the switching the pump on.
Comparisson of Dissolving Methods
Efficiency of Xenon dissolution by diffusion (a), by direct introduction through a glass frit (b) and through oxygenator membranes (c) were compared.
For most of the medical/biological relevant solvents method (b) is not feasible due to severe foaming.
|
|
|
|
|
(a) Diffusion |
(b) Bubbling |
(c) Membranes |
In Lipofundin and DMSO HP-Xenon can only be dissolved by the membrane method. While Lipofundin is a promising contrast agent, DMSO is used as a solvent for biological systems (e.g. proteins) and new experiments for structural analysis could become possible.
Solvents which form stable clathrates with Xe (e.g. ethanol [5]) are an exception. Here, directly introducing the gas (b) is by far the best way to dissolve Xenon.
|
|
|
First Images in a 1.5T clinical Scanner
The feasibility of the method was shown by imaging dissolved HP-Xe in Lipofundin on a clinical Siemens Magnetom 1.5T scanner. A setup similar to prototype was used (CELGARD® membranes).
Xe images accomplished by “shaking” and using the membrane approach to dissolve the gas were compared. SNR ratio of both methods are in the same order (~ 2.8 : 1).
Next experiments to get in vivo images in small animals (e.g. rats) are being planned.
|
|
|
|
|
|
|
A promising new method to dissolve hyperpolarized Xenon in several solvents with following properties was found:
• fast molecularly dissolution of gas
• no formation of foams or bubbles
• no observed loss of polarization
•
continuously operation possible even at high
pressures
A Comparison with other methods of Xenon dissolution by NMR and direct measurement of partial pressures showed the advantages of the membrane method. Increasing the efficiency of dissolution by applying pressure allowed the investigation of Xenon chemical shift in bicells.
Further improvement of the materials, modules and the method will lead to an apparatus for easy production of hyperpolarized Xenon solutions as a new class of contrast agents in MRI.
Additionally such hyperpolarized solvents permit new experiments in structure elucidation of complex biological systems, which are momentarily hampered by low concentrations and/or strong foaming.
The adsorbed Xenon can then function as a structural target and polarisation can be locally transferred to the adsorbing biomolecule to elucidate lipophilic sites.
