Feasibility of Simultaneous Measurements of GABA, Glx and GSH in the Thalamus Using Hadamard Encoding and Reconstruction of MEGA-Edited Spectroscopy (HERMES) – Application to TBI
Abstract Text: Introduction: Despite the lack of imaging findings, a significant fraction of mild TBI (mTBI) patients report persistent post-concussive symptoms (PCS). The thalamus may be uniquely affected in TBI as it serves as the central relay station for most of the sensory pathways to the cerebral cortex. Changes in the thalamic structure and functional connectivity have been observed following mTBI and correlated with PCS.
Such changes could be preceded by changes in neurotransmitter function. At the cellular level, rapid release of glutamate (Glu) following TBI may trigger up/down regulation of ionotropic receptor expression, leading to a Glu and GABA imbalance. Abnormal iron accumulation may also occur in the thalamus, causing oxidative stress. In vivo measurement of GABA and Glu, as well as glutathione (GSH), a major marker of oxidative stress, could therefore provide critical information regarding the underlying pathophysiological process. Recently developed Hadamard Encoding and Reconstruction of MEGA-Edited Spectroscopy (HERMES) can simultaneously provide information on these metabolites within a single acquisition.
In this study we first established the feasibility of HERMES in the thalamus on healthy volunteers. Further, we investigated the effectiveness of HERMES on mTBI subjects by comparing the spectral quality against control subjects.
Methods: HERMES was administered on four groups of subjects, including 3 groups of mTBI patients in the acute (N=10), subacute (N=6), and chronic (N=8) stages, and a control group (N=15) with no known brain injuries. Spectra were acquired on a Siemens PrismaFIT 3T scanner with TE=80ms, TR=2s, and 320 averages. A 3x3x2.5 cm3 voxel of interest covered the thalamus bilaterally. HERMES data were processed with frequency-and-phase correction using Gannet software. Metabolite levels were quantified following CSF-correction. The spectral quality, as measured by fitting errors and FWHM of individual metabolite peaks, and the metabolite levels were compared between each mTBI group against the control group using unpaired t-test.
Results: The mean fitting errors for the three metabolites from all the groups were below 12% with no significant difference between any of the mTBI group and the control group. No significant difference was found in the mean FWHM except for GABA between the acute mTBI and the control group (24.4±2.1 Hz vs 22.2±2.6 Hz). Overall, the fitting errors and FWHM results indicate robust spectral quality in the control and mTBI subjects.
GABA and Glx levels in the acute and chronic mTBI groups, respectively, were found to be significantly higher than the control group. The reliability of the metabolite levels as measured by coefficient of variation (CV) was generally comparable between the mTBI and control groups for GABA (19-26% vs 23%) and Glx (11-14% vs 16%). No significant difference was found in the GSH levels between the mTBI and control groups. The mTBI groups, however, showed higher CVs for GSH versus the control group (27-35% vs 20%).
Discussion: The larger GABA FWHM in the acute mTBI group could be explained by the increased area under the GABA peak. The trend for higher CVs for GSH among the mTBI groups may be attributed to the higher variability in the underlying pathophysiological process. The simultaneous increase of GABA and Glx in the thalamus in the acute group suggests an immediate imbalance in the inhibitory and excitatory pathways to facilitate normal tissue functioning. Elevated levels of Glx in the chronic group may suggest ongoing chronic inflammation leading to excitotoxic conditions in the thalamus. A continued long-term imbalance in the neurotransmitter function could potentially lead to thalamic atrophy with associated symptoms as have been observed in previous studies. The findings in this study should be treated with caution given the small sample size. We continue to accrue more patients into this study, and it remains to be seen if the preliminary findings presented here hold with a larger sample size.
Conclusion: In this study, we demonstrated the feasibility of simultaneous measurements of GABA, Glx and GSH using HERMES in the thalamus of the mTBI patients with data quality comparable to the control subjects. Preliminary results suggest imbalance between the excitatory and inhibitory pathways in the acute stage and a sustained excitatory state in the chronic stage.