Literature DB >> 35100003

Standalone portable xenon-129 hyperpolariser for multicentre clinical magnetic resonance imaging of the lungs.

Graham Norquay1, Guilhem J Collier1, Oliver I Rodgers1, Andrew B Gill2, Nicholas J Screaton2, Jim Wild1.   

Abstract

OBJECTIVES: Design and build a portable xenon-129 (129Xe) hyperpolariser for clinically accessible 129Xe lung MRI.
METHODS: The polariser system consists of six main functional components: (i) a laser diode array and optics; (ii) a B0 coil assembly; (iii) an oven containing an optical cell; (iv) NMR and optical spectrometers; (v) a gas-handling manifold; and (vi) a cryostat within a permanent magnet. All components run without external utilities such as compressed air or three-phase electricity, and require just three mains sockets for operation. The system can be manually transported in a lightweight van and rapidly installed on a small estates footprint in a hospital setting.
RESULTS: The polariser routinely provides polarised 129Xe for routine clinical lung MRI. To test the concept of portability and rapid deployment, it was transported 200 km, installed at a hospital with no previous experience with the technology and 129Xe MR images of a diagnostic quality were acquired the day after system transport and installation.
CONCLUSION: This portable 129Xe hyperpolariser system could form the basis of a cost-effective platform for wider clinical dissemination and multicentre evaluation of 129Xe lung MR imaging. ADVANCES IN KNOWLEDGE: Our work successfully demonstrates the feasibility of multicentre clinical 129Xe MRI with a portable hyperpolariser system.

Entities:  

Mesh:

Substances:

Year:  2022        PMID: 35100003      PMCID: PMC9153725          DOI: 10.1259/bjr.20210872

Source DB:  PubMed          Journal:  Br J Radiol        ISSN: 0007-1285            Impact factor:   3.629


Introduction

Magnetic resonance imaging (MRI) of hyperpolarised xenon-129 (129Xe) is an emerging research and clinical imaging modality, with demonstrated clinical impact in a range of lung pathologies such as cystic fibrosis, interstitial lung diseases, asthma and obstructive lung diseases.[1] Additionally, hyperpolarised 129Xe MRI has been demonstrated in other organs (e.g. the brain and kidneys) where dissolved-phase xenon can be used to study tissue perfusion and gas exchange.[2,3] While conventional proton (1H) MRI makes use of proton spin densities in tissues and blood in the lungs, owing to the comparatively lower density of 129Xe in the lung airspaces or dissolved in blood and other tissues, hyperpolarised 129Xe MRI relies on exogenous enhancement (“hyperpolarisation”) of the magnetisation of the 129Xe nuclei using laser polarisation techniques.[4] This enhancement leads to a five orders of magnitude MR signal enhancement for hyperpolarised 129Xe when compared to thermally polarised129Xe. This is critical in enabling MR images of 129Xe in the lung airspaces, as well as in well-perfused organs such as the kidneys[3] and the brain.[5,6] Early development of noble gas MRI of human lungs focussed on polarised helium-3 (3He) because of its isotopic purity and higher intrinsic MR signal.[7] In recent years, however, noble gas lung MRI research has moved from 3He towards 129Xe due to reasons of greater availability and thus lower cost of xenon, which can be extracted from the air.[8] For clinical applications of hyperpolarised 129Xe MRI, it is essential that xenon doses are generated rapidly (scan run time~10 min) with sufficient 129Xe polarisation levels (>20%) to enable clinical interpretation of 129Xe MR images. The most common technique used to magnetise 129Xe is spin-exchange optical pumping.[4] Two methods of spin-exchange optical pumping have emerged from physics research for clinical 129Xe MRI applications: batch mode[9,10] and continuous-flow mode.[11,12] The main operational difference between batch and continuous-flow mode is that batch mode does not require cryogenic accumulation of xenon, whereas continuous-flow mode does. This difference generally leads to two distinct performance regimes for batch and continuous-flow mode, namely (i) batch mode yields higher 129Xe polarisation (50%–90%) with slower xenon production rates (100–1000 mL/h) and (ii) continuous-flow mode yields lower 129Xe polarisation (20%–50%) but with faster production rates (1000–3500 mL/h). The faster production rates currently obtainable on continuous-flow hyperpolarisers offer increased potential for routine clinical 129Xe MRI when compared to batch mode systems given typical xenon throughput demands. For example, at the University of Sheffield, where the modality is being used clinically for lung disease assessment, the throughput is as many as five patients/day, each requiring between 1,000 mL and 3,500 mL of hyperpolarised 129Xe for a given MR scan protocol. While modern continuous-flow polarisers demonstrate the feasibility of clinical 129Xe MRI, widespread clinical uptake is still restricted by the need for additional lab space and/or the electrical and mechanical estates utilities necessary to install, house and run a commercially-available 129Xe polariser. Our solution was to design and build a portable 129Xe polariser for rapid low-cost deployment for clinical 129Xe lung MRI. The design is a self-contained, stand-alone and transportable device that can quickly be installed on-site or be taken temporarily to another facility where there is interest in using hyperpolarised 129Xe. Thus removing the need for costly stationary equipment installation and not interfering with existing infrastructure required for operation. In this work we build on previous work aimed at increasing portability of 129Xe polariser technology,[13,14] by developing a compact 129Xe hyperpolariser optimised for high-volume xenon production for clinical lung MRI in a hospital setting. The design criteria of the system were as follows: (i) readily transportable to an external clinical MRI facility; (ii) compact with no need for additional site infrastructure (i.e. it should run on mains electricity without the need for compressed air supplies), (iii) regulatory approved for in vivo use; and (iv) provide doses of 129Xe for high-quality clinical lung MRI in <20 min.

Methods and materials

Equipment

The polariser comprises the following components: (i) a laser diode array with beam shaping optics; (ii) an electromagnetic static B0 coil assembly; (iii) an oven containing a cylindrical optical cell; (iv) NMR and optical spectrometers; (v) a laser power meter; (vi) a gas-handling manifold; and (vii) a cryostat within a permanent magnetic field. See Figure 1 for a photo of the polariser with labelled components. The laser light for optical pumping is generated by an air-cooled laser diode array (Brightlock Ultra-100®, QPC Lasers, Sylmar, US) providing ~50 W of laser light tuned to the rubidium valence electron D1 line (795 nm). The class IV laser diode is coupled to an optical fibre to custom optics that produce left circularly polarised light in a homogeneous circular beam that is incident on the oven window. Stray laser light is mitigated by enclosing the beam, optics and oven with front and back plates, however laser safety goggles are worn as an additional precautionary measure.
Figure 1.

Photo of the assembled 129Xe polariser on the left and on the right a map of the 4-square coil geometry magnetic field coils. A is a retractable permanent magnet housing spiral glassware and an aluminium dewar for cryogenic accumulation of xenon. B is a 75W air-cooled laser diode and C is an optical train containing polarisation optics and a beam expander to output a 7.5-cm-diameter beam of circularly polarised light. D (i) and (ii) show the magnetic field coil placement and a simulation of the generated field lines (COMSOL Multiphysics). E is a ceramic forced-air oven housing a 40-cm-length cylindrical cell containing ~1 g droplet of rubidium metal. G is a computer workstation for on-board NMR of 129Xe in the cell and monitoring of the laser diode output. F shows the gas-handling manifold and H is a Tedlar bag for collection of sublimated hyperpolarised 129Xe gas.

Photo of the assembled 129Xe polariser on the left and on the right a map of the 4-square coil geometry magnetic field coils. A is a retractable permanent magnet housing spiral glassware and an aluminium dewar for cryogenic accumulation of xenon. B is a 75W air-cooled laser diode and C is an optical train containing polarisation optics and a beam expander to output a 7.5-cm-diameter beam of circularly polarised light. D (i) and (ii) show the magnetic field coil placement and a simulation of the generated field lines (COMSOL Multiphysics). E is a ceramic forced-air oven housing a 40-cm-length cylindrical cell containing ~1 g droplet of rubidium metal. G is a computer workstation for on-board NMR of 129Xe in the cell and monitoring of the laser diode output. F shows the gas-handling manifold and H is a Tedlar bag for collection of sublimated hyperpolarised 129Xe gas. The oven is made of compressed calcium silicate ceramic with antireflection glass windows and a cylindrical optical cell made from borosilicate glass with volume 1767 cm3 (40 cm length, 7.5 cm diameter) loaded with ~1 g rubidium. The oven is heated using a bespoke air compressor / heating element combination. The oven and cell are contained within a magnetic field generated by a four-coil electromagnet using square coils with a side-length of ~60 cm for optimised field homogeneity [Figure 1, D (ii)]. The nominal B0 field is maintained at 2.4 mT, corresponding to a 28 kHz resonance frequency for 129Xe nuclei. The gas manifold directs gas flow from a cylinder (located in a gas trolley ~1 m away from the system to prevent magnetic field distortion) containing 3% isotopically enriched xenon (86% 129Xe), 10% N2 and 87% He (~2000 L volume at 2000 psi pressure, BOC, Linde gases group). After passing through a gas getter and pressure regulator set for a cell pressure of 2 bar, the gas enters the optical cell via high-vacuum PTFE stop-cock valves. During continuous-flow operation, the xenon gas mixture is flowed through the cell in a direction counter to the incident laser light at a rate of 1000 sccm (standard cubic cm per minute), corresponding to a xenon flow rate of 1800 mL/h. The polarised gas exiting the cell is flowed through Tygon (Saint-Gobain, Coventry, UK) tubing towards spiral glassware. The glassware is held within a field of 250 mT from a NdBFe horseshoe permanent magnet and submerged in a dewar containing liquid N2, whereupon the xenon in the flowing gas mixture is cryogenically separated from He and N2, which are removed as exhaust gases through a vacuum line. The gas flow rate is controlled using a mass flow meter in line with a diaphragm vacuum pump which generates a downstream pressure of ~2 mbar. Once sufficient xenon snow is deposited in the spiral, it is defrosted by immersion in room-temperature water where the frozen xenon sublimates to the gas phase and flows into a Tedlar bag (Jensen Inert Products, Coral Springs, US) ready for QC release to the subject. In situ laser monitoring is provided by an optical spectrometer and a photon power meter. A home-built low-field nuclear magnetic resonance spectrometer using a LabVIEW software interface and a National Instruments data acquisition board (PCIe-6351 X-series multifunction DAQ) is used for in-system signal monitoring of the polarised 129Xe. The signals from 129Xe is acquired using a 100-turn surface B1 coil located on top of the optical cell inside the oven. The assembled 129Xe polariser (see photo in Figure 1) is standalone, occupies a footprint of 1.34m length x 0.72m width x 1.2m height, is powered by 3 × 240 V / 50 Hz AC mains sockets, weighs less than 150 kg and can be moved easily by two people into a lightweight van. The polariser and its associated GMP processes have been regulatory approved for the manufacture of polarised 129Xe for clinical lung MRI by the UK Medicines and Healthcare products Regulatory Agency (MHRA) under the Manufacturing Specials licence number MS-18739.

Lung imaging

To test the concept of portability of the polariser, it was transported 200 km in a van to a novice clinical radiology site with no previous in-house experience of 129Xe MRI. Lung imaging was performed using a transmit-receive vest coil (CMRS, US) on a healthy male volunteer the next day after 3 h of commissioning on site on a 1.5 T MR scanner (Siemens, MAGNETOM, Avanto). Prior to running the ventilation sequence, we calibrated the flip angle and centre frequency from an inhaled dose of 1L of a 3% xenon gas mixture. Sequence details are shown within the caption of Figure 2. The lung imaging work was conducted with full UK ethics approval (STH 20825, IRAS ref 265997) and UK regulatory manufacturing licence for xenon (MS-18739).
Figure 2.

A: 129Xe lung ventilation MR images acquired on a 1.5 T clinical scanner at our imaging centre from a subject with asthma. Sequence details: repetition time msec/echo time msec, 6.7/2.3; field of view, 400×400 mm; matrix size, 80×80; slice thickness, 10 mm; flip angle, 10°; and bandwidth, 200 Hz/pixel. B: 129Xe lung ventilation MR images acquired on a 1.5 T clinical scanner at an external imaging centre the day after polariser transportation. Sequence details: volunteer, healthy male; repetition time msec/echo time msec, 6.9/2.3; field of view, 367×420 mm; matrix size, 56×64; slice thickness, 22 mm; flip angle, 10°; and bandwidth, 200 Hz/pixel.

A: 129Xe lung ventilation MR images acquired on a 1.5 T clinical scanner at our imaging centre from a subject with asthma. Sequence details: repetition time msec/echo time msec, 6.7/2.3; field of view, 400×400 mm; matrix size, 80×80; slice thickness, 10 mm; flip angle, 10°; and bandwidth, 200 Hz/pixel. B: 129Xe lung ventilation MR images acquired on a 1.5 T clinical scanner at an external imaging centre the day after polariser transportation. Sequence details: volunteer, healthy male; repetition time msec/echo time msec, 6.9/2.3; field of view, 367×420 mm; matrix size, 56×64; slice thickness, 22 mm; flip angle, 10°; and bandwidth, 200 Hz/pixel.

Results

Figure 2A shows MR images of 129Xe acquired from the lungs of patients with the polariser in routine use at our MR imaging centre; inhaled dose 500 mL of xenon topped up to 1 litre with medical grade nitrogen acquired on a GE 450W 1.5 T MRI scanner housed in a relocatable mobile unit. Figure 2B shows 129Xe lung images from the healthy volunteer at the external imaging centre; inhaled dose 300 mL of xenon topped up to 1 litre with medical grade nitrogen. The 129Xe polarisation was measured to be ~30% (using the method described in ref.[15]), and the 300 mL and 500 mL doses were generated in 10 min and 17 min (xenon production rate of 1800 mL/h). The signal-to-noise ratio over all image slices was measured to be >40.

Discussion

We have demonstrated the feasibility of a lightweight, high-volume production rate, portable 129Xe polariser for hyperpolarised gas lung MRI imaging. Portability of the system was demonstrated with a rapid turnkey installation at an external clinical imaging centre. From an installation and running perspective, the system fits easily in the back of a transit van, operates reliably on three 240V/50 Hz mains electricity sockets, and the on-board air compressor and heater is able to heat the oven to 140°C in ~20 min with <1°C fluctuation once thermal stability is reached. The system has been tested in various locations including in the close confines and noisy electrical environment of the equipment room of a relocatable mobile MRI scanner. The only additional requirements at the installation site are completion of risk assessments and also access to liquid nitrogen; this should not pose an impediment with respect to accessibility to other clinical MR centres, as all large hospitals should have access to liquid nitrogen storage. In order to compare operational performance against other polariser systems, we employ the metric dose equivalence rate (DER),[16] a metric that couples the quality and quantity of xenon gas available for patient inhalation within a given time period – i.e. it is a product of the operational 129Xe polarisation and the xenon volume production rate. It is quantified numerically as DER = f×P×Q, where f is the isotopic fraction of 129Xe (normalised to 1.00 in the calculation), P is the fractional 129Xe polarisation and Q is the xenon volume production rate.[16] Using this formula with a Q of 1800 mL/h and P of 0.30 for the system described here yields a DER equal to 540 mL/h, which enables acquisition of diagnostic-quality MR lung images using 500 mL of xenon gas produced in <20 min. The performance of our portable polariser is approaching that of a permanent clinical 129Xe polariser system, which has a DER value of ~1000 mL/h and is used routinely for diagnostic clinical lung imaging.[12] It is worth noting that to facilitate multicentre applications of 129Xe MRI, as well as increasing a site’s accessibility to polariser technology, it is important to standardise xenon dosing and imaging protocols as outlined in ref.[17] and to also consider the additional outlay on 129Xe RF coils and broadband RF amplifiers on the scanner itself.

Conclusions

In this work, we have demonstrated a self-contained, portable polariser that rapidly generates highly polarised 129Xe gas. The system routinely provides high-SNR lung imaging, and can easily be transported and installed at novice clinical imaging centres where no specialist equipment is required for polariser installation. The system should form the basis of a cost-effective platform for wider dissemination and multicentre clinical evaluation of hyperpolarised 129Xe MR imaging. Increased access to 129Xe hyperpolarisation technology outlined in this work should create new opportunities to explore and utilise the clinical impact of 129Xe MRI for diagnosis, prognosis, and treatment response monitoring of a number of diseases within the lungs and beyond.
  14 in total

1.  Near-unity nuclear polarization with an open-source 129Xe hyperpolarizer for NMR and MRI.

Authors:  Panayiotis Nikolaou; Aaron M Coffey; Laura L Walkup; Brogan M Gust; Nicholas Whiting; Hayley Newton; Scott Barcus; Iga Muradyan; Mikayel Dabaghyan; Gregory D Moroz; Matthew S Rosen; Samuel Patz; Michael J Barlow; Eduard Y Chekmenev; Boyd M Goodson
Journal:  Proc Natl Acad Sci U S A       Date:  2013-08-14       Impact factor: 11.205

2.  Dose and pulse sequence considerations for hyperpolarized (129)Xe ventilation MRI.

Authors:  Mu He; Scott H Robertson; S Sivaram Kaushik; Matthew S Freeman; Rohan S Virgincar; John Davies; Jane Stiles; William M Foster; H Page McAdams; Bastiaan Driehuys
Journal:  Magn Reson Imaging       Date:  2015-04-30       Impact factor: 2.546

3.  Batch-Mode Clinical-Scale Optical Hyperpolarization of Xenon-129 Using an Aluminum Jacket with Rapid Temperature Ramping.

Authors:  Jonathan R Birchall; Panayiotis Nikolaou; Aaron M Coffey; Bryce E Kidd; Megan Murphy; Michael Molway; Liana B Bales; Boyd M Goodson; Robert K Irwin; Michael J Barlow; Eduard Y Chekmenev
Journal:  Anal Chem       Date:  2020-03-02       Impact factor: 6.986

4.  Imaging Human Brain Perfusion with Inhaled Hyperpolarized 129Xe MR Imaging.

Authors:  Madhwesha R Rao; Neil J Stewart; Paul D Griffiths; Graham Norquay; Jim M Wild
Journal:  Radiology       Date:  2017-08-31       Impact factor: 11.105

5.  Note: Spin-exchange optical pumping in a van.

Authors:  C Chauvin; L Liagre; C Boutin; E Mari; E Léonce; G Carret; B Coltrinari; P Berthault
Journal:  Rev Sci Instrum       Date:  2016-01       Impact factor: 1.523

6.  Protocols for multi-site trials using hyperpolarized 129 Xe MRI for imaging of ventilation, alveolar-airspace size, and gas exchange: A position paper from the 129 Xe MRI clinical trials consortium.

Authors:  Peter J Niedbalski; Chase S Hall; Mario Castro; Rachel L Eddy; Jonathan H Rayment; Sarah Svenningsen; Grace Parraga; Brandon Zanette; Giles E Santyr; Robert P Thomen; Neil J Stewart; Guilhem J Collier; Ho-Fung Chan; Jim M Wild; Sean B Fain; G Wilson Miller; Jaime F Mata; John P Mugler; Bastiaan Driehuys; Matthew M Willmering; Zackary I Cleveland; Jason C Woods
Journal:  Magn Reson Med       Date:  2021-09-03       Impact factor: 4.668

7.  Dissolved hyperpolarized xenon-129 MRI in human kidneys.

Authors:  Jorge Chacon-Caldera; Adam Maunder; Madhwesha Rao; Graham Norquay; Oliver I Rodgers; Matthew Clemence; Claudio Puddu; Lothar R Schad; Jim M Wild
Journal:  Magn Reson Med       Date:  2019-08-09       Impact factor: 4.668

8.  Measuring 129 Xe transfer across the blood-brain barrier using MR spectroscopy.

Authors:  Madhwesha R Rao; Graham Norquay; Neil J Stewart; Jim M Wild
Journal:  Magn Reson Med       Date:  2021-01-17       Impact factor: 4.668

9.  ^{129}Xe-Rb Spin-Exchange Optical Pumping with High Photon Efficiency.

Authors:  G Norquay; G J Collier; M Rao; N J Stewart; J M Wild
Journal:  Phys Rev Lett       Date:  2018-10-12       Impact factor: 9.161

View more
  1 in total

1.  BJR functional imaging of the lung special feature: introductory editorial.

Authors:  Philippe A Grenier; Eric A Hoffman; Nicholas Screaton; Joon Beom Seo
Journal:  Br J Radiol       Date:  2022-04       Impact factor: 3.629

  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.