1
Simultaneous measurement of GABA and GSH using HERMES at 3T Kennedy Krieger Institute Unlocking Potential Introduction Methods Muhammad G. Saleh 1,2 , Georg Oeltzschner 1,2 , Kimberly L. Chan 1,2,3 , Nicolaas A. J. Puts 1,2 , Mark Mikkelsen 1,2 , Michael Schar 1 , Richard A. E. Edden 1,2 1. Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins University School of Medicine. 2. F. M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute; 3. Department of Biomedical Engineering, The Johns Hopkins University School of Medicine • Simulated, phantom and in vivo HERMES spectra show: 1) Excellent segregation of edited signals, and 2) Strong agreement with separately acquired MEGA-PRESS data • In practice, HERMES allows simultaneous editing of both GABA and GSH within the same duration and with equivalent SNR as a single MEGA-PRESS measurement • Incorporating editing lobes at 1.5 ppm into the OFF GABA scans would result in a macromolecular-suppressed GABA-edited experiment 4 1. Mescher et al., NMR Biomed. 1998;11:266-272. 2. Chan et al., Magn Reson Med. 2016;76:11-19. 3. Simpson et al., Magn Reson Med. 2017;77:23-33. 4. Henry PG et al. Magn Reson Med. 2001; 45:517–20. • GABA is the main inhibitory neurotransmitter in the human brain • GSH (glutathione) is the most abundant redox compound in the brain, serving an important role in minimizing damage caused by reactive oxygen species • Spectral editing of the MR spectrum (Fig. 1) allows selective detection of low-concentration compounds, such as GABA 1 and GSH Pitfalls: 1) Selectively edits one metabolite at a time and from one brain region 2) Requires relatively long acquisition times within the time constraints of an MR examination AIM: Hadamard Editing and Reconstruction of MEGA-edited Spectroscopy (HERMES) 2 to simultaneously edit GABA and GSH • Requires four sub-experiments (A, B, C, D), which can be combined to form GABA- and GSH-edited spectra (Fig. 2) • Hadamard encoding treats each metabolite orthogonally, allowing for independent editing without signal cross-talk between the metabolite spectra Fig. 2: HERMES editing of GSH and GABA. a) Inversion profiles of editing pulses in the four sub-experiments A-D. b) Voxel location. c) Dual-frequency inversion targets two metabolites simultaneously (Experiment A) and a Sinc-Gaussian pulse is used to invert a single metabolite. d) Hadamard transformation of the sub-experiments yields the separate GSH- and GABA-edited spectra. HERMES editing of GABA and GSH This work was supported in part by NIH grants NIH R01 016909 and P41 015089 Fig. 1: (a) Mescher-Garwood (MEGA) difference editing of GABA. Frequency-selective pulses (red arrow) selectively target GABA spins at 1.9 ppm. Subtraction remove ovelapping signals and reveal signals impacted by the editing pulses. (b) Interleaved 2-step (ON and OFF) MEGA editing scheme: A typical acquisition comprises of 320 averages requiring 11-min of scan time. 3 2 4 5 ON GABA OFF GABA Experiment A Experiment B ON GABA OFF GABA Experiment A Experiment B . . . ppm b 1 2 3 4 ppm EDIT PULSE ON EDIT PULSE OFF DIFFERENCEx10 GABA Cr Cho NAA a • Three phantoms: • 4 mM GSH • 4 mM GSH + 4 mM GABA • 10 mM GABA • TR/TE 2000/80 ms • 27-mL isotropic voxel Phantom • HERMES & MEGA-PRESS • 10 healthy adult subjects • TR/TE 2000/80 ms • 320 averages • 47 mL isotropic voxel • ~11 min per acquisition In vivo • FID Application 3 • Voxel-center only • Shaped refocusing and editing pulses • 20-ms (62-Hz FWHM) editing pulses Simulations Results Fig. 3: Phantom/simulated spectra (black/red). a) HERMES experiments A-D for GSH & GABA. b) Hadamard reconstructed spectra show separation of GSH- and GABA-edited signals. c) MEGA-PRESS spectra of each metabolite show the same edited lineshapes. d) HERMES spectra of a mixed phantom demonstrate successful simultaneous editing. OFF ON ON OFF ON ON OFF OFF Experiment C Experiment B Experiment A Experiment D GABA GSH Hadamard Reconstruction MEGA-PRESS A-B+C-D A+B-C-D HERMES of GSH + GABA phantom b a c d GSH GABA HERMES MEGA HERMES MEGA ppm 1 2 3 4 ppm 1 2 3 4 Fig. 5: In vivo HERMES and MEGA-PRESS spectra from all subjects. Simultaneously acquired HERMES spectra are shown in orange and sequentially acquired MEGA-PRESS spectra are overlaid in each case in blue. There is excellent agreement in terms of data quality, SNR and quantitative measures. Discussion References Acknowledgments Fig. 4: In vivo HERMES of a typical subject. a) The separate sub-spectra A-D are plotted. The saturation range of the editing pulses is shown on each spectrum as a grayscale overlay. b) HERMES spectra show a GSH-edited spectrum in the A-B+C-D combination and a GABA-edited spectrum in the A+B-C-D combination. Experiment A Experiment C Experiment D Experiment B 0 1 2 3 4 5 OFF OFF OFF OFF ON ON ON ON GSH GABA ppm ppm 1 2 3 4 A-B+C-D A+B-C-D a b a OFF GABA ON GSH ON GABA OFF GSH ON GABA ON GSH OFF GABA OFF GSH 3 2 ppm 4 5 Experiment C Experiment B Experiment A Experiment D +1 -1 Inversion +1 -1 Inversion +1 -1 Inversion +1 -1 Inversion d = A+B-C-D A-B+C-D GSH-edited Spectrum GABA-edited Spectrum c b Sinc-Gaussian for single-lobe inversion Cosine-Sinc-Gaussian for dual-lobe inversion
