INTRODUCTION: Functional lung imaging is a substantial unaddressed medical need. Hyperpolarized (HP) ¹²⁹Xe gas has been shown to be promising for clinical-research pulmonary imaging. Since the first in-vivo MRI demonstration in the 1990s, much of the effort in the field of HP ¹²⁹Xe gas has been focused on the technologies of ¹²⁹Xe gas hyperpolarization and on demonstrating the effectiveness of this imaging modality for detection and characterization of a wide range of pulmonary diseases. Despite its demonstrated efficacy in clinical studies, HP ¹²⁹Xe MRI has not yet enjoyed clinical translation anywhere in the world due to a major critical translational roadblock: high cost and complexity of clinical hyperpolarization equipment and imaging. Clinical MRI scanners are equipped with proton transmission-detection capability only and cannot perform the detection of HP ¹²⁹Xe. We propose addressing this unmet medical need through the use of proton-hyperpolarized propane gas. HP propane gas is cheap, and it can be readily produced and imaged using clinical MRI scanners. We have demonstrated the feasibility of clinical-scale production of HP propane previously². Here, we report on recent advances in development of a handheld low-cost hyperpolarizer that can be readily disposed after an imaging examination.

METHODS: We employ a 250 mL commercially available aluminum spray can for temporary storage of pressurized mixture of propylene:parahydrogen (1:1 and 1:4) gases at ~9 atm total pressure. We have employed 98.5% parahydrogen prepared using ultra-high purity hydrogen gas⁴. The disposable storage container is equipped with high-flow valve (>0.5 standard liters per second flow). The storage container shown in Figure 2 is mated to the disposable reactor made of food-grade copper tubing filled with 100 mg of 1% Rh/TiO₂ catalyst and copper beads. The beads are added to improve thermal management². Once the valve is actuated, the compressed gas mixture is released through the mini-reactor resulting in production of hyperpolarized propane gas stream. The performance of the hyperpolarizer was monitored using 1.4 T bench-top NMR spectrometer running dynamic pseudo-2D NMR, Figures 2 and 3. Custom MATLAB code³ was employed for all spectroscopic data processing and determination of relaxation rates and signal enhancement values. All reported imaging studies were performed using a 3.0 T MRI scanner and 2D EPI, GRE and FIESTA pulse sequences. The acquisition parameters are provided in corresponding Figure captions. The photograph of the experimental setup employed for imaging studies is shown in Figure 4.

RESULTS AND DISCUSSION: Parahydrogen gas¹ serves as a source of hyperpolarization (Figure 1) in parahydrogen induced polarization of propane gas via pairwise addition of parahydrogen. The utility of a a commercial aluminum spray can as a temporary storage vessel was evaluated by monitoring the decay of the gas mixture potency over time: parahydrogen decays with the exponential rate constant of 6.0+/-0.5 days, Figure 2e. We conclude that this decay rate is sufficient for temporary storage of the gas mixture and overnight shipping.
Systematic studies of flow rate, flow duration, and reaction pressure revealed that the production of HP propane gas is very robust (Figures 2 and 3) leading to apparent polarization values of 1-1.5% as detected at 1.4 T—in line with our previous studies^2,5 using a benchtop polarizer. Although the polarization levels are substantially lower than those typically obtained for HP ¹²⁹Xe, we note that each HP molecule carries double the polarization load (two HP protons versus one HP ¹²⁹Xe atom), and each proton is typically 3.6 times more sensitive than ¹²⁹Xe at the same nominal polarization. Moreover, protons have nearly 100% natural abundance versus 26% natural abundance of ¹²⁹Xe. These compounding factors make 1.5%-hyperpolarized propane equivalent to 43%-hyperpolarized ¹²⁹Xe – this math does not even factor in higher detection frequency of protons and highly-optimized RF coils already available in the clinical practice, which additionally favor HP propane imaging.
For 3T imaging studies, we employed a simplified setup without reactor heating as shown in Figure 4. We note that the valves are introduced only for the purpose of the phantom studies and will be obviated for future in vivo studies. Feasibility, of 2D GRE imaging is shown in Figure 1. Feasibility of multi-slice sub-second 2D EPI is demonstrated in Figure 5 --- three selected slices are shown from the stack of 8 images. Multiple 2D images were obtained from the same batch of HP propane gas, allowing one to monitor gas injection in the phantom and the hyperpolarization decay. We note that this technology will additionally benefit from low-field MRI^5,6, which allows protecting the long-lived spin states of HP propane gas--therefore allowing for a substantially longer time window of gas preparation, administration, and imaging.

CONCLUSION: We conclude that a parahydrogen:propylene mixture can be successfully stored for days in a disposable handheld container that can be potentially shipped. Robust performance of the disposable handheld hyperpolarizer was demonstrated in vitro. The feasibility of ultra-fast 2D EPI, GRE, and FIESTA imaging of HP propane has been demonstrated. This pilot study bodes well for near-future in vivo validation of this prospective inhalable contrast agent for functional lung imaging applications. All-in-all, the new instrumentation presented here enabled low-cost on-demand production of HP propane using a handheld device, and imaging using conventional clinical MRI scanners.