Methodological consensus on clinical proton MRS of the brain: Review and recommendations

The Goldilocks Zone for 3-T MRS Studies Using Semi-LASER: Determining the Optimal Balance Between Repetition Time and Scan Time

1H-MR spectroscopy studies often use a short TR to reduce scan time. However, this causes significant T1-weighting (T1w), which can alter metabolite estimates due to acquisition factors rather than biochemistry. Our goal was to determine the optimal balance between scan time and TR that minimizes T1w effects in semi-LASER MRS at 3 T. Spectra were acquired in the posterior cingulate cortex of five healthy volunteers (2 male/3 female, mean age 25 ± 2 years) and analyzed using FSL-MRS. The SNR and metabolite estimates of five metabolites were compared at TR = 2, 5, and 8 s, under conditions of "similar scan time" with varying acquisition numbers and "constant number of acquisitions" with varying scan times. T1 relaxation times derived from metabolite estimates were compared to literature. With a 25% longer scan time, SNRTR = 5s was 34% higher and SNRTR = 2s was 29% higher than SNRTR = 8s for data with "similar scan times." Using a TR = 5 s or longer, the SNR per minute is consistent for metabolites with T1s less than 2 s. Metabolite estimate trends were similar for the two different scenarios of "similar scan time" and "same number of acquisitions," where all metabolite estimates were obtained without metabolite T1 correction. The largest metabolite estimates were found at TR = 8 s, they were 10-15% lower at TR = 5 s and 15-30% lower at TR = 2 s. T1 values agreed with literature values. At TR = 2 s, SNR per minute and metabolite estimates were lower due to reduced signal availability via T1w effects. TR = 8 s had the least amount of T1w effects, but results in lower SNR per minute. TR = 5 s had enough signal recovery to be robust to T1w effects, and yielded the largest SNR for similar scan times, with a clinically feasible scan time of 5 m 40 s. Using semi-LASER MRS with a TR = 5 s is recommended to improve the sensitivity of MRS to changes in metabolite estimates.

Neurobiological differences in early-onset obsessive-compulsive disorder: A study of the glutamatergic system based on functional magnetic resonance spectroscopy

In this study, the combination of functional state magnetic resonance spectroscopy (fMRS) and cognitive tasks was used to conduct subgroup analyses on early-onset OCD (EO) and non-early-onset OCD (non-EO) and explore differences in the glutamatergic system and cognitive function among OCD subtypes. A total of 70 OCD and 30 healthy controls (HCs) underwent clinical evaluation and were subsequently divided into the EO or non-EO groups. Next, both resting and functional state MRS data were collected, with the anterior cingulate cortex (ACC) serving as the region of interest. Quantitative analysis of MRS data yielded precise neurometabolic concentrations, which were then statistically analyzed alongside inhibitory function, as measured by the Go-nogo task. The final analysis included 92 participants (22 EO-OCD, 41 non-EO OCD, and 29 HCs). EO-OCD patients had significantly higher Glx levels (p = 0.044) and lower GSH levels (p = 0.009) in the functional state compared to the non-EO group. Moreover, in the EO group, correlation analysis revealed a positive correlation between the functional state Glx levels and the average response time for errors in the nogo task (r = 0.526, p = 0.014). Additionally, resting-state GSH levels were positively correlated with total Y-BOCS scores (r = 0.854, p < 0.001). Overall, early-onset OCD may represent a distinct subtype that requires targeted interventions, as evidenced by the imbalance in the glutamatergic system observed in early-onset OCD patients. Additionally, in early-onset patients, Glx concentration during activation was related to cognitive impairment.

Magnetic resonance spectroscopy and the menstrual cycle: A multi-centre assessment of menstrual cycle effects on GABA & GSH

BACKGROUND

Gamma-aminobutyric acid (GABA) and glutathione (GSH) play a significant role in the functioning of a healthy brain and can both be quantified using magnetic resonance spectroscopy (MRS). Several small-scale studies have suggested MRS measured GABA may fluctuate with the menstrual cycle, but the effects on GSH are unknown. Utilising recent developments in MRS acquisition, this multi-lab study explores this issue across 4 distinctive brain regions.

NEW METHODS

Data were analysed from 12 independent sites from which a total of 30 women were scanned during three phases of their menstrual cycle corresponding to early follicular, ovulation and mid luteal phases. HERMES and HERCULES sequences were used to measure GABA and GSH in voxels located in the left motor cortex, left posterior insular, medial parietal and medial frontal. Linear mixed models were used to assess the variability contributed by site, participant and menstrual cycle phase.

RESULTS

Similar variance was attributed to site and menstrual cycle phase for both GABA and GSH data. No systematic changes in GABA or GSH were revealed for any voxel as a consequence of menstrual cycle phase.

COMPARISON WITH EXISTING METHODS

Despite our larger sample size and inclusion of more brain regions we fail to replicate previous findings of GABA change as a consequence of menstrual cycle phase. We also show for the first time that MRS measures of GSH so not significantly alter with cycle.

CONCLUSIONS

Our findings suggest that the menstrual cycle has minimal impact on MRS measures of GABA and GSH. The presence of a menstrual cycle should not be used as justification for exclusion of women in MRS studies.

Diagnostic Accuracy of 1H-MRS Using PRESS and MEGA-PRESS Techniques in the Preoperative Grading of Patients With Gliomas

BACKGROUND

Edited MRS technique such as MEshcher-GArwood Point RESolved Spectroscopy (MEGA-PRESS) can determine isocitrate dehydrogenase mutation (IDH) mutation status in patients with gliomas but its accuracy in assessing glioma grade has not yet been formally evaluated.

PURPOSE

To evaluate the diagnostic accuracy of metabolites such as lactate obtained from the PRESS and MEGA-PRESS sequences in the preoperative grading of glioma. To assess the prognostic value of those metabolite ratios in the overall survival of patients with gliomas.

STUDY TYPE

Prospective.

SUBJECTS

Sixty-nine subjects with gliomas (16 grade 2, 21 grade 3, and 32 grade 4). Mean age was 50.5 ± 16.7 years; 38 were male and 31 were female.

FIELD STRENGTH/SEQUENCE

3 T/MEGA-PRESS, PRESS.

ASSESSMENT

Single voxel PRESS and MEGA-PRESS spectra were obtained from tumors in patients undergoing preoperative MRI. Several tumor metabolites were measured from the PRESS, MEGA-PRESS edit-off, and difference spectra using LCModel (Linear Combination of Model Spectra) software. Diagnosis and glioma grading was done using the World Health Organization (WHO) 2016 classification. Overall survival was assessed.

STATISTICAL TESTS

Diagnostic accuracy was measured using receiver-operating characteristic (ROC) curve. Univariate and multivariate Cox proportional hazards modeling was used for the assessment of prognostic factors for time to death.

RESULTS

In the differentiation between low- vs. high-grade gliomas, tCr/tCho ratios obtained from PRESS and MEGA-PRESS sequences had similar accuracies (area under the ROC curves [AUCs] = 0.71) while Lac/NAA from PRESS had a lower accuracy (AUC = 0.65). The presence of a detectable 2-hydroxyglutarate peak on the difference spectrum was a favorable prognostic factor in univariate analysis (hazard ratio = 0.25, 95% confidence interval: 0.08-0.83). No other metabolite was found to be a significant prognostic factor in univariate and multivariate analyses.

DATA CONCLUSION

Edited MRS can be used to detect metabolites which can help in the preoperative grading of gliomas and in determination of the overall survival. A separate PRESS acquisition is needed for lactate quantification.

PLAIN LANGUAGE SUMMARY

Gliomas are brain tumors that vary in severity. This study explored the use of two advanced MR spectroscopy techniques (PRESS and MEGA-PRESS) in detecting tumor metabolites. The authors found that both techniques' choline/creatine ratio showed moderate accuracy in identifying high-grade gliomas. Lactate was better revealed with the PRESS technique and was associated with high-grade gliomas. They confirmed that the MEGA-PRESS technique allowed additional detection of 2-hydroxyglutarate in IDH-mutant gliomas, which was linked to better survival. These findings emphasize that advanced MR spectroscopy can extract metabolic information time-efficiently, which can be used to improve the preoperative diagnosis of patients with gliomas.

LEVEL OF EVIDENCE

1 TECHNICAL EFFICACY: Stage 2.

Neurometabolite Alterations Associated With Cannabis Use: A Proton Magnetic Resonance Spectroscopy Meta-Analysis

Little is known about the neurometabolic effects of cannabis use. Using meta-analytic modeling of proton magnetic resonance spectroscopy (1H-MRS) studies, this study aimed to assess the differences in brain metabolite levels associated with cannabis use (PROSPERO: CRD42020209890) to inform treatment development for cannabis use disorder (CUD). Hedge's g with random-effects modeling was used, and heterogeneity and publication bias indices were assessed. A complete literature search was conducted, and 15 studies met the inclusion criteria (e.g., 1H-MRS, cannabis group compared to a control group, brain region-specific results, necessary data to complete modeling). There were 29 models across gray matter regions in the brain. All models had between 2 and 5 studies (k), indicating that results should be interpreted with caution due to the limited number of available studies. Compared to the control groups, the cannabis-using groups showed lower levels of GABA and N-acetylaspartate in the anterior cingulate cortex (k = 3); lower glutamate in the basal ganglia/striatum (k = 2); and lower glutamine and myo-inositol in the thalamus (k = 2; although the two effect sizes came from the same sample). This is the first meta-analysis to consolidate the extant 1H-MRS studies focused on the neurometabolic effects of cannabis. Despite the few studies available, the evidence suggests cannabis use may impact important neural processes, including glutamatergic and GABAergic functioning (glutamate, glutamine, and GABA), neural health (N-acetylaspartate), and glial functioning (myo-inositol). The findings should be interpreted with caution considering the small sample size; the inability to test the impact of demographic, substance use, and methodological factors; and the heterogeneity of studies. Understanding the neurobiological effects of cannabis may inspire novel pharmacotherapy and/or psychosocial interventions for CUD.

Reproducibility Made Easy: A Tool for Methodological Transparency and Efficient Standardized Reporting Based on the Proposed MRSinMRS Consensus

Recent expert consensus publications have highlighted the issue of poor reproducibility in magnetic resonance spectroscopy (MRS) studies, mainly due to the lack of standardized reporting criteria, which affects their clinical applicability. To combat this, guidelines for minimum reporting standards (MRSinMRS) were introduced to aid journal editors and reviewers in ensuring the comprehensive documentation of essential MRS study parameters. Despite these efforts, the implementation of MRSinMRS standards has been slow, attributed to the diverse nomenclature used by different vendors, the variety of raw MRS data formats, and the absence of appropriate software tools for identifying and reporting necessary parameters. To overcome this obstacle, we have developed the REproducibility Made Easy (REMY) standalone toolbox. REMY software supports a range of MRS data formats from major vendors like GE (.7), Siemens (.ima, .rda, .dcm), Philips (.spar/.sdat), and Bruker (.method), and MRS-NIfTI (.nii/nii.gz/.json) files facilitating easy data import and export through a user-friendly interface. REMY employs external libraries such as spec2nii and pymapVBVD to accurately read and process these diverse data formats, translating complex header information into a comprehensive structure that adheres to consensus reporting standards, thereby ensuring compatibility and ease of use for researchers in generating reproducible MRS research outputs. Users can select and import datasets, choose the appropriate vendor and data format, and then generate an MRSinMRS table, log file, and methodological documents in both Latex and PDF formats by just clicking one button. No coding knowledge is required, making the tool accessible to a wider range of users, including researchers and clinicians without programming expertise. This eliminates technical challenges related to data formatting and reporting. REMY effectively populated key sections of the MRSinMRS table with data from all supported file types. Accurate generation of hardware parameters including field strength, manufacturer, and scanner software version were demonstrated. However, it could not input data for RF coil and additional hardware information due to their absence in the files. For the acquisition section, REMY accurately read and populated fields for the pulse sequence name, nominal voxel size, repetition time (TR), echo time (TE), number of acquisitions/excitations/shots, spectral width (Hz), and number of spectral points, significantly contributing to the completion of the "Acquisition" fields of the table. Furthermore, REMY generates a boilerplate methods text section for manuscripts. The use of REMY will facilitate more widespread adoption of the MRSinMRS checklist within the MRS community, making it easier to write and report acquisition parameters effectively.

Mapping GABA+/Glx in experimental temporal lobe epilepsy using edited-MRSI at 9.4T

Mesial temporal lobe epilepsy (MTLE) is a common drug-resistant epilepsy, often requiring surgery to remove the epileptogenic zone (EZ). The mean efficiency of the surgery is 50-70 %, so the accurate spatial localization of the EZ remains a challenge. In a previous study from Hamelin et al. (2021), gamma-aminobutyric acid (GABA), the main inhibitory neurotransmitter, was shown to be highly increased in the EZ of a MTLE mouse model, with a concomitant decrease of Glx (Glutamate + glutamine). The authors proposed the GABA/Glx ratio as a potential specific biomarker of the EZ. As it is the only way to measure GABA and Glx non-invasively in vivo, we propose to use Magnetic Resonance Spectroscopy (MRS) methods to improve the non-invasive localization of the EZ. Our study introduces an original GABA-edited Magnetic Resonance Spectroscopic Imaging (MRSI) method, i.e. the MEGA-LASER CSI, and the dedicated data processing pipeline, to map GABA and Glx in pre-clinical settings. The MEGA-LASER CSI sequence was validated using multi-compartment phantoms with varying GABA concentrations, demonstrating a high accuracy to spatially discriminate the different compartments. In vivo experiments revealed a significant increase of the GABA+/Glx ratio in the EZ of epileptic animals (n=30) compared to SHAM ones (n=15), strongly correlated with ex vivo data. Preliminary immunohistochemistry experiments revealed astrocytic localization of GAD65 in the EZ, suggesting a shift in GABA synthesis from nerve endings to astrocytes. Phantom and in vivo experiments prove that our original workflow for GABA and Glx mapping is suitable for use in neuroscience and pre-clinical applications.

Proton Magnetic Resonance Spectroscopy for Pediatric Neuroimaging: Key Concepts for Practice

Magnetic resonance spectroscopy (MRS) of the brain provides the clinician an in vivo neurochemical assessment within the clinical magnetic resonance imaging setting. This information can yield specificity when addressing questions pertaining to brain health and metabolism while characterizing disease and injury, evaluating treatment response, and prognosticating outcome. Proton MRS techniques can be useful in narrowing the diagnostic differential and capturing time-sensitive information for the continually developing pediatric brain. This paper provides a review of key proton MRS topics relevant for usage in pediatric populations. We discuss magnetic field strength, pediatric-sized head coils, water suppression techniques, localization pulse sequences, post-processing methods, analysis, and interpretation. These elements all require special consideration, particularly for the immature brain. We introduce the fundamentals of spectral editing. Finally, we present illustrative examples employing proton MRS in clinical practice to begin to synthesize these concepts into practical application.

Altered visual cortex excitatory/inhibitory ratio following transient congenital visual deprivation in humans

Non-human animal models have indicated that the ratio of excitation to inhibition (E/I) in neural circuits is experience dependent, and changes across development. Here, we assessed 3T Magnetic Resonance Spectroscopy (MRS) and electroencephalography (EEG) markers of cortical E/I ratio in 10 individuals who had been treated for dense bilateral congenital cataracts, after an average of 12 years of blindness, to test for dependence of the E/I ratio on early visual experience in humans. First, participants underwent MRS scanning at rest with their eyes open and eyes closed, to obtain visual cortex Gamma-Aminobutyric Acid (GABA+) concentration, Glutamate/Glutamine (Glx) concentration, and the concentration ratio of Glx/GABA+, as measures of inhibition, excitation, and E/I ratio, respectively. Subsequently, EEG was recorded to assess aperiodic activity (1-20 Hz) as a neurophysiological measure of the cortical E/I ratio, during rest with eyes open and eyes closed, and during flickering stimulation. Across conditions, congenital cataract-reversal individuals demonstrated a significantly lower visual cortex Glx/GABA+ ratio, and a higher intercept and steeper aperiodic slope at occipital electrodes, compared to age-matched sighted controls. In the congenital cataract-reversal group, a lower Glx/GABA+ ratio was associated with better visual acuity, and Glx concentration correlated positively with the aperiodic intercept in the conditions with visual input. We speculate that these findings result from an increased E/I ratio of the visual cortex as a consequence of congenital blindness, which might require commensurately increased inhibition in order to balance the additional excitation from restored visual input. The lower E/I ratio in congenital cataract-reversal individuals would thus be a consequence of homeostatic plasticity.

The periaqueductal gray in chronic low back pain: dysregulated neurotransmitters and function

ABSTRACT

Mechanisms underlying chronic pain are insufficiently understood, hampering effective treatment approaches. Preclinical evidence suggests a potential contribution of decreased excitatory (glutamatergic) and increased inhibitory (γ-aminobutyric acid [GABA]ergic) neurotransmission in the periaqueductal gray (PAG), a key descending pain modulatory brainstem area. This magnetic resonance spectroscopy (MRS) study investigated (1) whether a lower excitatory/inhibitory balance is also observed in the PAG of patients with nonspecific chronic low back pain (CLBP) and (2) whether the excitatory/inhibitory balance relates to psychophysical measures of descending pain modulation and pain sensitivity. Magnetic resonance spectroscopy was acquired on a 3T MR system in 41 patients with CLBP and 29 age- and sex-matched controls. Descending pain modulation and pain sensitivity were evaluated using conditioned pain modulation and pressure pain stimuli, respectively, which were both assessed at the lower back as the most painful area and the nondominant hand as a pain-free, remote area. Patients with CLBP presented with a lower glutamate + glutamine (Glx)/GABA ratio compared with controls (P = 0.002), driven by both decreased Glx (P = 0.012) and increased GABA (P = 0.038). Controls with lower Glx/GABA were more sensitive to pressure pain in both areas, but this association was missing in the patients (lower back: P = 0.004; hand: P = 0.002). Patients with more severe clinical pain showed impaired descending pain modulation at the hand (P = 0.003). In line with preclinical evidence, these findings support a dysregulated PAG in patients with CLBP that might be associated with dysfunctional descending pain inhibition.

A longitudinal study of hippocampal subfield volumes and hippocampal glutamate levels in antipsychotic-naïve first episode psychosis patients

BACKGROUND

Previous studies have implicated hippocampal abnormalities in the neuropathology of psychosis spectrum disorders. Reduced hippocampal volume has been reported across all illness stages, and this atrophy has been hypothesized to be the result of glutamatergic excess. To test this hypothesis, we measured hippocampal subfield volumes and hippocampal glutamate levels in antipsychotic naïve first episode psychosis patients (FEP) and the progression of volume decline and changes in glutamate levels over a 16-week antipsychotic drug (APD) trial. We aimed to determine if subfield volumes at baseline were associated with glutamate levels, and if baseline glutamate levels were predictive of change in subfield volumes over time.

METHODS

We enrolled ninety-three medication-naïve FEP participants and 80 matched healthy controls (HC). T1 and T2 weighted images and magnetic resonance spectroscopy (MRS) data from a voxel prescribed in the left hippocampus were collected from participants at baseline and after 6 and 16 weeks of APD treatment. Hippocampal subfield volumes were assessed using FreeSurfer 7.1.1., while glutamate levels were quantified using jMRUI version 6.0. Data were analyzed using linear mixed models.

RESULTS

We found regional subfield volume deficits in the CA1, and presubiculum in FEP at baseline, that further expanded to include the molecular and granule cell layer of the dentate gyrus (GC/ML/DG) and CA4 by week 16. Baseline hippocampal glutamate levels in FEP were not significantly different than those of HC, and there was no effect of treatment on glutamate. Glutamate levels were not related to initial subfield volumes or volume changes over 16 weeks.

CONCLUSION

We report a progressive loss of hippocampal subfield volumes over a period of 16 weeks after initiation of treatment, suggestive of early progression in neuropathology. Our results do not suggest a role for glutamate as a driving factor. This study underscores the need to further research the mechanism(s) underlying this phenomenon as it has implications for early intervention to preserve cognitive decline in FEP participants.

Alterations in gamma-aminobutyric acid and glutamate neurotransmission linked to intermittent theta-burst stimulation in depression: a sham-controlled study

Gamma-aminobutyric acid (GABA) and glutamate are implicated in the antidepressant effects of repetitive transcranial magnetic stimulation (rTMS), though findings from magnetic resonance spectroscopy (MRS) are inconsistent. Furthermore, the relationship between GABAA-receptor availability and rTMS outcomes remains largely unexplored. In this study, GABA and glutamate levels in the dorsal anterior cingulate cortex (dACC) were measured using a 1H-MRS MEGA-PRESS sequence in 42 patients with bipolar or unipolar depression, both before and after a sham-controlled, double-blind clinical trial involving intermittent theta-burst stimulation (iTBS) over the dorsomedial prefrontal cortex. A subset of 28 patients also underwent [11C]flumazenil positron emission tomography (PET) to measure whole-brain GABAA-receptor availability and mean receptor availability in the nucleus accumbens and dACC. Depressive symptoms were assessed using the self-rated Montgomery Åsberg Depression Rating Scale (MADRS-S). The results indicated no significant changes in neurotransmitter levels or GABAA-receptor availability post-iTBS in either the active or sham conditions. However, changes in MADRS-S scores after active iTBS were positively correlated with changes in GABA levels in the dACC (r(13) = 0.54, p = 0.04) and baseline GABAA-receptor availability in the nucleus accumbens (r(11) = 0.66, p = 0.02). These correlations were absent in the sham group. The findings suggest that a reduction in GABA within targeted frontostriatal circuits can be part of the antidepressant mechanism of iTBS, challenging previous research. Additionally, they indicate a potential predictive role for frontostriatal GABAA-receptor availability in the treatment of depression using dorsomedial prefrontal iTBS.

Optimized Long-TE 1H sLASER MR Spectroscopic Imaging at 3T for Separate Quantification of Glutamate and Glutamine in Glioma

BACKGROUND

Glutamate and glutamine are critical metabolites in gliomas, each serving distinct roles in tumor biology. Separate quantification of these metabolites using in vivo MR spectroscopy (MRS) at clinical field strengths (≤ 3T) is hindered by their molecular similarity, resulting in overlapping, hence indistinguishable, spectral peaks.

PURPOSE

To develop an MRS imaging (MRSI) protocol to map glutamate and glutamine separately at 3T within clinically feasible time, using J-modulation to enhance spectral differentiation, demonstrate its reliability/reproducibility, and quantify the metabolites in glioma subregions.

STUDY TYPE

Prospective.

POPULATION

Phantoms, 5 healthy subjects, and 30 patients with suspected glioma. IDH wild-type glioblastoma cases were evaluated to establish a uniform group.

FIELD STRENGTH/SEQUENCE

3T, Research protocol: 2D 1H sLASER MRSI (40 and 120 ms TE) with water reference, 3D T1-weighted and 2D T2-weighted. Trial-screening process: T1-weighted, T1-weighted contrast-enhanced, T2-weighted, FLAIR.

ASSESSMENT

Spectral simulations and phantom measurements were performed to design and validate the protocol. Spectral quality/fitting parameters for scan-rescan measurements were obtained using LCModel. The proposed long-TE data were compared with short-TE data. BraTS Toolkit was employed for fully automated tumor segmentation.

STATISTICAL TESTS

Scan-rescan comparison was performed using Bland-Altman analysis. LCModel coefficient of modeling covariance (CMC) between glutamate and glutamine was mapped to evaluate their model interactions for each spectral fitting. Metabolite levels in tumor subregions were compared using one-way ANOVA and Kruskal-Wallis. A p value < 0.05 was considered statistically significant.

RESULTS

Spectral quality/fitting parameters and metabolite levels were highly consistent between scan-rescan measurements. A negative association between glutamate and glutamine models at short TE (CMC = -0.16 ± 0.06) was eliminated at long TE (0.01 ± 0.05). Low glutamate in tumor subregions (non-enhancing-tumor-core: 5.35 ± 4.45 mM, surrounding-non-enhancing-FLAIR-hyperintensity: 7.39 ± 2.62 mM, and enhancing-tumor: 7.60 ± 4.16 mM) was found compared to contralateral (10.84 ± 2.94 mM), whereas glutamine was higher in surrounding-non-enhancing-FLAIR-hyperintensity (9.17 ± 6.84 mM) and enhancing-tumor (7.20 ± 4.42 mM), but not in non-enhancing-tumor-core (4.92 ± 3.38 mM, p = 0.18) compared to contralateral (2.94 ± 1.35 mM).

DATA CONCLUSION

The proposed MRSI protocol (~12 min) enables separate mapping of glutamate and glutamine reliably along with other MRS-detectable standard metabolites in glioma subregions at 3T.

EVIDENCE LEVEL

1 TECHNICAL EFFICACY: Stage 3.

WALINET: A water and lipid identification convolutional neural network for nuisance signal removal in 1 H $$ {}^1\mathrm{H} $$ MR spectroscopic imaging

PURPOSE

Proton magnetic resonance spectroscopic imaging (  1 H $$ {}^1\mathrm{H} $$ -MRSI) provides noninvasive spectral-spatial mapping of metabolism. However, long-standing problems in whole-brain  1 H $$ {}^1\mathrm{H} $$ -MRSI are spectral overlap of metabolite peaks with large lipid signal from scalp, and overwhelming water signal that distorts spectra. Fast and effective methods are needed for high-resolution  1 H $$ {}^1\mathrm{H} $$ -MRSI to accurately remove lipid and water signals while preserving the metabolite signal. The potential of supervised neural networks for this task remains unexplored, despite their success for other MRSI processing.

METHODS

We introduce a deep learning method based on a modified Y-NET network for water and lipid removal in whole-brain  1 H $$ {}^1\mathrm{H} $$ -MRSI. The WALINET (WAter and LIpid neural NETwork) was compared with conventional methods such as the state-of-the-art lipid L2 regularization and Hankel-Lanczos singular value decomposition (HLSVD) water suppression. Methods were evaluated on simulated models and in vivo whole-brain MRSI using NMRSE, SNR, CRLB, and FWHM metrics.

RESULTS

WALINET is significantly faster and needs 8s for high-resolution whole-brain MRSI, compared with 42min for conventional HLSVD+L2. WALINET suppresses lipid and water in the brain by 25-45 and 34-53-fold, respectively. WALINET has better performance than HLSVD+L2, providing: (1) more lipid removal with 41% lower NRMSE; (2) better metabolite signal preservation with 71% lower NRMSE in simulated data; 155% higher SNR and 50% lower CRLB in in vivo data. Metabolic maps obtained by WALINET in healthy subjects and patients show better gray-/white-matter contrast with more visible structural details.

CONCLUSIONS

WALINET has superior performance for nuisance signal removal and metabolite quantification on whole-brain  1 H $$ {}^1\mathrm{H} $$ -MRSI compared with conventional state-of-the-art techniques. This represents a new application of deep learning for MRSI processing, with potential for automated high-throughput workflow.

Creatine monohydrate pilot in Alzheimer's: Feasibility, brain creatine, and cognition

BACKGROUND

Preclinical studies suggest that creatine monohydrate (CrM) improves cognition and Alzheimer's disease (AD) biomarkers. However, there is currently no clinical evidence demonstrating the effects of CrM in patients with AD.

METHODS

In this single-arm pilot trial, we investigated the feasibility of 20 g/day CrM for 8 weeks in 20 patients with AD. We measured compliance throughout; serum creatine at baseline, 4 weeks, and 8 weeks; and brain total creatine (tCr) and cognition (National Institutes of Health [NIH] Toolbox, Mini-Mental State Examination [MMSE]) at baseline and 8 weeks.

RESULTS

Nineteen participants achieved the target of ≥80% compliance with the CrM intervention. Serum Cr was elevated at 4 and 8 weeks (p < .001) and brain tCr increased by 11% (p < .001). Cognition improved on global (p = .02) and fluid (p = .004) composites, List Sorting (p = .001), Oral Reading (p < .001), and Flanker (p = .05) tests.

DISCUSSION

Our data suggest that CrM supplementation is feasible in AD and provides preliminary evidence for future efficacy and mechanism studies.

Trial Registration

ClinicalTrials.gov, NCT05383833, registered on May 20, 2022.

Highlights

Creatine monohydrate supplementation was feasible in patients with Alzheimer's disease.Creatine monohydrate was associated with increased brain total creatine.Creatine monohydrate was associated with improvements in cognition.Efficacy of creatine monohydrate in Alzheimer's disease should be studied further.

Pitfalls in 2HG detection with TE-optimized MRS at 3T

Background and Purpose

In-vivo magnetic resonance spectroscopy (MRS) of 2-hydroxyglutarate (2HG) may provide diagnostic and monitoring biomarkers in isocitrate dehydrogenase (IDH)-mutated glioma. A previous meta-analysis has shown good diagnostic accuracy of TE-optimized PRESS for IDH-mutated glioma, but most studies feature IDH-wildtype glioma as a comparison. However, when considering newly identified brain lesions that may mimic glioma, full characterization of its diagnostic utility should also consider the accuracy of 2HG measurement in non-tumor tissue. Therefore, we tested how well TE-optimized 2HG levels distinguish between IDH-mutated glioma and non-tumor tissue, in this case, normal-appearing brain. We further examined the impact of different spectral modeling strategies (baseline stiffness, macromolecule inclusion, and basis set composition).

Materials and Methods

48 patients with diagnosed/suspected IDH-mutated glioma were enrolled. 3T MRS data were acquired from tumor and contralateral non-tumor tissue with PRESS localization (TE = 97 ms, optimized for 2HG detection) and analyzed with 'LCModel' software. Receiver operating characteristic analysis evaluated 2HG estimates' ability to distinguish IDH-mutated glioma from non-tumor brain tissue. Modeling interactions between 2HG and other metabolites were evaluated to identify reasons for potential false-positive 2HG detection.

Results

TE-optimized PRESS distinguished IDH-mutated glioma from non-tumor tissue with lower sensitivity (range 0.76-0.62) and specificity (0.85-0.78) than literature suggests for IDH-mutated vs. IDH-wildtype glioma. Strong negative correlations between gamma-aminobutyric acid (GABA) and 2HG persisted across all modeling strategies and may lead to false-positive 2HG detection in non-tumor tissue. We further present a cautionary example from a patient on a ketogenic diet, showing that the ketone body acetone can interfere with 2HG detection.

Conclusions

Spectral overlap with GABA and acetone can lead to false-positive 2HG detection in non-tumor tissue. Clinicians need to be mindful of these pitfalls when interpreting 2HG estimates.

Deep neural network modeling for brain tumor classification using magnetic resonance spectroscopic imaging

This study is driven by the complex and specialized nature of magnetic resonance spectroscopy imaging (MRSI) data processing, particularly within the scope of brain tumor assessments. Traditional methods often involve intricate manual procedures that demand considerable expertise. In response, we investigate the application of deep neural networks directly to raw MRSI data in the time domain. Given the significant health risks associated with brain tumors, the necessity for early and accurate detection is crucial for effective treatment. While conventional MRI techniques encounter limitations in the rapid and precise spatial evaluation of diffuse gliomas, both accuracy and efficiency are often compromised. MRSI presents a promising alternative by providing detailed insights into tissue chemical composition and metabolic changes. Our proposed model, which utilizes deep neural networks, is specifically designed for the analysis and classification of spectral time series data. Trained on a dataset that includes both synthetic and real MRSI data from brain tumor patients, the model aims to distinguish MRSI voxels that indicate pathological conditions from healthy ones. Our findings demonstrate the model's robustness in classifying glioma-related MRSI voxels from those of healthy tissue, achieving an area under the receiver operating characteristic curve of 0.95. Overall, these results highlight the potential of deep learning approaches to harness raw MR data for clinical applications, signaling a transformative impact on diagnostic and prognostic assessments in brain tumor examinations. Ongoing research is focused on validating these approaches across larger datasets, to establish standardized guidelines and enhance their clinical utility.

Test-retest precision of brain metabolites in healthy participants using 31P-MRS and 1H MEGA-PRESS on a 3T multi-nuclear MRI system

Background

Magnetic resonance spectroscopy (MRS) enables the non-invasive quantification of brain metabolites, and its reliability is crucial for accurate interpretation of disease state. This study assessed the test-retest precision of phosphorus-31 (31P)-MRS and hydrogen (1H)-MEscher-GArwood Point RESolved Spectroscopy (MEGA-PRESS) in measuring 31P metabolites, γ-aminobutyric acid (GABA), and glutathione (GSH) using a 3T multi-nucleus magnetic resonance imaging (MRI) system.

Methods

In total, 32 participants, who underwent two scanning sessions within three days, using two dimensional (2D)-chemical shift imaging (CSI)-31P-MRS and 1H-MEGA-PRESS sequences, were enrolled in the study. γ-aminobutyric acid and macromolecules (GABA+), glutamate and glutamine (Glx), GSH, and 12 31P metabolites were analyzed using the MATLAB-based tool Gannet and jMRUI software. Precision was assessed based on the coefficients of variation (CVs) and Bland-Altman plots.

Results

The results revealed that potential of hydrogen (pH) and phosphocreatine (PCr) showed the greatest stability as evidenced by low CVs, suggesting reliable measurements across sessions. The adenosine triphosphates (ATPs) showed considerable stability. Conversely, metabolites, such as phosphomonoesters (PMEs) and phosphodiesters (PDEs), located to the left of PCr, showed reduced stability, while glycerophosphatidylcholine (GPTC) had the highest CV, indicating significant variability in clinical practice. Among the various brain regions, intermediate areas such as the temporal lobe and thalamus exhibited greater stability than peripheral regions such as the frontal and occipital lobes. Single-voxel MEGA-PRESS measurements showed that Glx and GABA+ had higher precision than GSH.

Conclusions

Both the 31P-MRS and 1H-MEGA-PRESS sequences showed high precision in measuring brain metabolites, but some metabolites showed higher stability than others. These results are crucial for exploring the clinical and research applications of these methods, and provide a solid foundation for subsequent investigations.

Current and prospective roles of magnetic resonance imaging in mild traumatic brain injury

There is unmet clinical need for biomarkers to predict recovery or the development of long-term sequelae of mild traumatic brain injury, a highly prevalent condition causing a constellation of disabling symptoms. A substantial proportion of patients live with long-lasting sequelae affecting their quality of life and ability to work. At present, symptoms can be assessed through clinical tests; however, there are no imaging or laboratory tests fully reflective of pathophysiology routinely used by clinicians to characterize post-concussive symptoms. Magnetic resonance imaging has potential to link subtle pathophysiological alterations to clinical outcomes. Here, we review the state of the art of MRI research in adults with mild traumatic brain injury and provide recommendations to facilitate transition into clinical practice. Studies utilizing MRI can inform on pathophysiology of mild traumatic brain injury. They suggest presence of early cytotoxic and vasogenic oedema. They also show that mild traumatic brain injury results in cellular injury and microbleeds affecting the integrity of myelin and white matter tracts, all processes that appear to induce delayed vascular reactions and functional changes. Crucially, correlates between MRI parameters and post-concussive symptoms are emerging. Clinical sequences such as T1-weighted MRI, susceptibility-weighted MRI or fluid attenuation inversion recovery could be easily implementable in clinical practice, but are not sufficient, in isolation for prognostication. Diffusion sequences have shown promises and, although in need of analysis standardization, are a research priority. Lastly, arterial spin labelling is emerging as a high-utility research as it could become useful to assess delayed neurovascular response and possible long-term symptoms.

The relationship between optical coherence tomography and magnetic resonance imaging measurements in people with multiple sclerosis: A systematic review and meta-analysis

BACKGROUND

Several studies show that optical coherence tomography (OCT) metrics e with cognition, disability, and brain structure in people with multiple sclerosis (PwMS). This review the correlation between OCT parameters and magnetic resonance imaging (MRI) measurements in PwMS.

METHODS

A comprehensive search of PubMed/MEDLINE, Embase, Scopus, and Web of Science was performed, including studies published in English up to November 29, 2024 to identify studies reporting quantitative data on the correlation between baseline OCT parameters and MRI measurements in PwMS. The meta-analysis was performed using R software version 4.4.0.

RESULTS

From 4931 studies, 68 studies on 6168 PwMS (67.4 % female) were included. The most significant correlations were found between peripapillary retinal nerve fiber layer (pRNFL) thickness and lower T1 lesion volume r = -0.42 (95 % CI: -0.52 to -0.31, p-value <0.001, I2 = 24 %), greater thalamic volume r = 0.39 (95 % CI: 0.17 to 0.61, p-value <0.001, I2 = 81 %), and lower T2 lesion volume r = -0.37 (95 % CI: -0.54 to -0.21, p-value <0.001, I2 = 85 %), respectively. Additionally, lower macular ganglion cell-inner plexiform layer (mGCIPL) thickness showed the most significant correlations with positive and lower thalamic volume r = 0.37 (95 % CI: 0.1 to 0.64, p-value = 0.008, I2 = 88 %), and positive and lower grey matter volume (GMV) 0.33 (95 % CI: 0.15 to 0.52, p-value <0.001, I2 = 81 %), respectively.

CONCLUSION

pRNFL and mGCIPL thickness are correlated with MRI measurements, suggesting that OCT can serve as a non-invasive, cost-effective, and complementary tool to MRI for enhancing the exploring of brain structural changes in PwMS.

Myo-inositol in the Dorsal Anterior Cingulate Cortex is Associated with Anxiety-to-Eat in Anorexia Nervosa

Anorexia nervosa (AN) is a mental and behavioral health condition characterized by an intense fear of body weight or fat gain, restriction of food intake resulting in low body weight, and distorted body image. Substantial research has focused on general anxiety in AN, but less is known about eating-related anxiety and its underlying neurobiological mechanisms. We sought to characterize anxiety-to-eat in AN and to examine neurometabolite levels in the dorsal anterior cingulate cortex (dACC), a brain region putatively involved in modulating anxiety-related responses, using edited magnetic resonance spectroscopy. Sixteen women hospitalized with AN and 16 women of healthy weight without a lifetime history of an eating disorder (healthy controls; HC) completed a computer-based behavioral task assessing anxiety-to-eat in response to images of higher (HED) and lower energy density (LED) foods. The AN group reported greater anxiety to eat HED and LED foods relative to the HC group. Both groups reported greater anxiety to eat HED foods relative to LED foods. The neurometabolite myo-inositol (myo-I) was lower in the dACC in AN relative to HC. In the AN group only, myo-I levels negatively predicted anxiety to eat HED but not LED foods and was independent of body mass index, duration of illness, and general anxiety. These findings provide new insight into the clinically challenging feature of eating-related anxiety in AN, and indicate potential for myo-I levels in the dACC to serve as a novel biomarker of illness severity or therapeutic target in individuals vulnerable to AN.

The Relationship Among Range Adaptation, Social Anhedonia, and Social Functioning: A Combined Magnetic Resonance Spectroscopy and Resting-State fMRI Study

BACKGROUND AND HYPOTHESIS

Social anhedonia is a core feature of schizotypy and correlates significantly with social functioning and range adaptation. Range adaptation refers to representing a stimulus value based on its relative position in the range of pre-experienced values. This study aimed to examine the resting-state neural correlates of range adaptation and its associations with social anhedonia and social functioning.

STUDY DESIGN

In study 1, 60 participants completed resting-state magnetic resonance spectroscopy and fMRI scans. Range adaptation was assessed by a valid effort-based decision-making paradigm. Self-reported questionnaires was used to measure social anhedonia and social functioning. Study 2 utilized 26 pairs of participants with high (HSoA) and low levels of social anhedonia (LSoA) to examine the group difference in range adaptation's neural correlates and its relationship with social anhedonia and social functioning. An independent sample of 40 pairs of HSoA and LSoA was used to verify the findings.

STUDY RESULTS

Study 1 showed that range adaptation correlated with excitation-inhibition balance (EIB) and ventral prefrontal cortex (vPFC) functional connectivity, which in turn correlating positively with social functioning. Range adaptation was specifically determined by the EIB via mediation of ventral-medial prefrontal cortex functional connectivities. Study 2 found HSoA and LSoA participants exhibiting comparable EIB and vPFC connectivities. However, EIB and vPFC connectivities were negatively correlated with social anhedonia and social functioning in HSoA participants.

CONCLUSIONS

EIB and vPFC functional connectivity is putative neural correlates for range adaptation. Such neural correlates are associated with social anhedonia and social functioning.

Brain Metabolite Differences in Fetuses With Cytomegalovirus Infection: A Magnetic Resonance Spectroscopy Study

BACKGROUND

Cytomegalovirus (CMV) is the most common intrauterine infection and may be associated with unfavorable outcomes. While some CMV-infected fetuses may show gross or subtle brain abnormalities on MRI, their clinical significance may be unclear. Conversely, normal development cannot be guaranteed in CMV-infected fetuses with normal MRI.

PURPOSE

To assess brain metabolite differences in CMV-infected fetuses using magnetic resonance spectroscopy (MRS).

STUDY TYPE

Retrospective.

SUBJECTS

Out of a cohort of 149 cases, 44 with maternal CMV infection, amniocentesis results, and good-quality MRS were included. CMV-infected fetuses with positive polymerase chain reaction (PCR) (N = 35) were divided based on MRI results as follows: typical brain abnormalities (gross findings, N = 8), exclusive white matter hyperintense signal (WMHS) on T2-weighted images (subtle findings, N = 7), and normal MRI (N = 20). Uninfected fetuses (negative PCR) with normal MRI were included as controls (N = 9).

FIELD STRENGTH

3 T, T2-weighted half Fourier single-shot turbo spin-echo (HASTE), T2-weighted true fast imaging with steady-state free precession (TrueFISP), T1- and T2*-weighted fast low angle shot (FLASH), and 1H-MRS single-voxel point resolved spectroscopy (PRESS) sequences.

ASSESSMENT

MRI findings were assessed by three radiologists, and metabolic ratios within the basal ganglia were calculated using LCModel.

STATISTICAL TESTS

Analysis of covariance test with Bonferroni correction for multiple comparisons was used to compare metabolic ratios between groups while accounting for gestational age. A P-value <0.05 was deemed significant.

RESULTS

MRS was successfully acquired in 63% of fetuses. Substantial agreement was observed between radiologists (Fleiss' kappa [k] = 0.8). Infected fetuses with gross MRI findings exhibited significantly reduced tNAA/tCr ratios (0.64 ± 0.08) compared with infected fetuses with subtle MRI findings (0.85 ± 0.19), infected fetuses with normal MRI (0.8 ± 0.14) and controls (0.81 ± 0.15). No other significant differences were detected (P ≥ 0.261).

CONCLUSION

Reduced tNAA/tCr within the apparently normal brain tissue was detected in CMV-infected fetuses with gross brain abnormalities, suggesting extensive brain damage. In CMV-infected fetuses with isolated WMHS, no damage was detected by MRS.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 3.

synMARSS-An End-To-End Platform for the Parametric Generation of Synthetic In Vivo Magnetic Resonance Spectra

Synthetic magnetic resonance spectra (MRS) are mathematically generated spectra which can be used to investigate the assumptions of data analysis strategies, optimize experimental design, and as training data for the development and validation of machine learning tools. In this work, we extend Magnetic Resonance Spectrum Simulator (MARSS), a popular MRS basis set simulation tool, to be able to generate synthetic spectra for an arbitrary MRS sequence. The extension, referred to as synMARSS, converts a basis set as well as a set of NMR, tissue-related and additional sequence parameters into high-quality synthetic spectra via a parametric model. synMARSS is highly versatile, incorporating T1 and T2 relaxation, arbitrary line shape distortions and diffusion, while also quickly generating the large amount of training data needed for machine learning applications. Additionally, we extend MARSS to non-1H nuclei, such as 2H, 13C, and 31P. We use synthetic spectra to investigate the effects of approximating 14N heteronuclear coupling as weak homonuclear coupling, which was found to have small effects on the quantified concentrations for major metabolites for the implementation of PRESS at short echo time, but these effects increased at longer echo times.

Intra- and inter-session reliability and repeatability of 1H magnetic resonance spectroscopy for determining total creatine concentrations in multiple brain regions

Using proton magnetic resonance spectroscopy (1H MRS) to determine total creatine (tCr) concentrations will become increasingly prevalent, as the role of creatine (Cr) in supporting brain health gains interest. Methodological limitations and margins of error in repeated 1H MRS, which often surpass reported effects of supplementation, permeate existing literature. We examined the intra- and inter-session reliability and repeatability of 1H MRS for determining tCr concentrations across multiple brain regions (midbrain, visual cortex and frontal cortex). Eighteen healthy adults aged 20-32 years were recruited (50% female; n = 14 intra-session; n = 15 inter-session). 1H Magnetic resonance imaging and spectroscopy were completed at 3 T. Intra-session analyses involved repeated 1H MRS of the midbrain, visual cortex and frontal cortex without participant or voxel repositioning, whereas inter-session analyses involved measurements of the same regions, but with participant and voxel repositioning between repeated measurements. The 1H MRS data (174 spectra) were analysed using TARQUIN and OSPREY, and voxel fractions (grey/white matter and CSF) were determined using segmentation. Our findings show that tCr concentrations can be determined reliably and repeatably using 1H MRS, within an error of <2%, and that large inter-regional differences in tCr concentration are present in the human brain. We provide new minimum detectable change data for tCr concentrations, a detailed discussion of the inherent error sources in repeated 1H MRS, including the substantial effect of the analysis package on tCr quantification, and suggestions for how these should be managed to improve the interpretability and clinical value of future research. More studies are needed to determine whether our findings can be replicated in other centres and different populations.

Metabolite T2 relaxation times decrease across the adult lifespan in a large multi-site cohort

PURPOSE

Relaxation correction is crucial for accurately estimating metabolite concentrations measured using in vivo MRS. However, the majority of MRS quantification routines assume that relaxation values remain constant across the lifespan, despite prior evidence of T2 changes with aging for multiple of the major metabolites. Here, we comprehensively investigate correlations between T2 and age in a large, multi-site cohort.

METHODS

We recruited approximately 10 male and 10 female participants from each decade of life: 18-29, 30-39, 40-49, 50-59, and 60+ y old (n = 101 total). We collected PRESS data at eight TEs (30, 50, 74, 101, 135, 179, 241, and 350 ms) from voxels placed in white-matter-rich centrum semiovale (CSO) and gray-matter-rich posterior cingulate cortex (PCC). We quantified metabolite amplitudes using Osprey and fit exponential decay curves to estimate T2.

RESULTS

Older age was correlated with shorter T2 for tNAA2.0, tCr3.0, tCr3.9, tCho, and tissue water (CSO and PCC), as well as mI and Glx (PCC only); rs = -0.22 to -0.63, all p < 0.05, false discovery rate (FDR)-corrected. These associations largely remained statistically significant when controlling for cortical atrophy. By region, T2 values were longer in the CSO for tNAA2.0, tCr3.9, Glx, and tissue water and longer in the PCC for tCho and mI. T2 did not differ by region for tCr3.0.

CONCLUSION

These findings underscore the importance of considering metabolite T2 differences with aging in MRS quantification. We suggest that future 3T work utilize the equations presented here to estimate age-specific T2 values instead of relying on uniform default values.

A data-driven algorithm to determine 1H-MRS basis set composition

Purpose

Metabolite amplitude estimates derived from linear combination modeling of MR spectra depend upon the precise list of constituent metabolite basis functions used (the "basis set"). The absence of clear consensus on the "ideal" composition or objective criteria to determine the suitability of a particular basis set contributes to the poor reproducibility of MRS. In this proof-of-concept study, we demonstrate a novel, data-driven approach for deciding the basis-set composition using Bayesian information criteria (BIC).

Methods

We have developed an algorithm that iteratively adds metabolites to the basis set using iterative modeling, informed by BIC scores. We investigated two quantitative "stopping conditions", referred to as max-BIC and zero-amplitude, and whether to optimize the selection of basis set on a per-spectrum basis or at the group level. The algorithm was tested using two groups of synthetic in-vivo-like spectra representing healthy brain and tumor spectra, respectively, and the derived basis sets (and metabolite amplitude estimates) were compared to the ground truth.

Results

All derived basis sets correctly identified high-concentration metabolites and provided reasonable fits of the spectra. At the single-spectrum level, the two stopping conditions derived the underlying basis set with 77-87% accuracy. When optimizing across a group, basis set determination accuracy improved to 84-92%.

Conclusion

Data-driven determination of the basis set composition is feasible. With refinement, this approach could provide a valuable data-driven way to derive or refine basis sets, reducing the operator bias of MRS analyses, enhancing the objectivity of quantitative analyses, and increasing the clinical viability of MRS.

Exploring in vivo human brain metabolism at 10.5 T: Initial insights from MR spectroscopic imaging

INTRODUCTION

Ultra-high-field magnetic resonance (MR) systems (7 T and 9.4 T) offer the ability to probe human brain metabolism with enhanced precision. Here, we present the preliminary findings from 3D MR spectroscopic imaging (MRSI) of the human brain conducted with the world's first 10.5 T whole-body MR system.

METHODS

Employing a custom-built 16-channel transmit and 80-channel receive MR coil at 10.5 T, we conducted MRSI acquisitions in six healthy volunteers to map metabolic compounds in the human cerebrum in vivo. Three MRSI protocols with different matrix sizes and scan times (4.4 × 4.4 × 4.4 mm³: 10 min, 3.4 × 3.4 × 3.4 mm³: 15 min, and 2.75×2.75×2.75 mm³: 25 min) were tested. Concentric ring trajectories were utilized for time-efficient encoding of a spherical 3D k-space with ∼4 kHz spectral bandwidth. B0/B1 shimming was performed based on respective field mapping sequences and anatomical T1-weighted MRI were obtained.

RESULTS

By combining the benefits of an ultra-high-field system with the advantages of free-induction-decay (FID-)MRSI, we present the first metabolic maps acquired at 10.5 T in the healthy human brain at both high (voxel size of 4.4³ mm³) and ultra-high (voxel size of 2.75³ mm³) isotropic spatial resolutions. Maps of 13 metabolic compounds (aspartate, choline compounds and creatine + phosphocreatine, γ-aminobutyric acid (GABA), glucose, glutamine, glutamate, glutathione, myo-inositol, scyllo-inositol, N-acetylaspartate (NAA), N-acetylaspartylglutamate (NAAG), taurine) and macromolecules were obtained individually. The spectral quality was outstanding in the parietal and occipital lobes, but lower in other brain regions such as the temporal and frontal lobes. The average total NAA (tNAA = NAA + NAAG) signal-to-noise ratio over the whole volume of interest was 12.1± 8.9 and the full width at half maximum of tNAA was 24.7± 9.6 Hz for the 2.75 × 2.75 × 2.75 mm³ resolution. The need for an increased spectral bandwidth in combination with spatio-spectral encoding imposed significant challenges on the gradient system, but the FID approach proved very robust to field inhomogeneities of ∆B0 = 45 ± 38 Hz (frequency offset ± spatial STD) and B1+ = 65 ± 11° within the MRSI volume of interest.

DISCUSSION

These preliminary findings highlight the potential of 10.5 T MRSI as a powerful imaging tool for probing cerebral metabolism. By providing unprecedented spatial and spectral resolution, this technology could offer a unique view into the metabolic intricacies of the human brain, but further technical developments will be necessary to optimize data quality and fully leverage the capabilities of 10.5 T MRSI.

Changes in glutamate levels in anterior cingulate cortex following 16 weeks of antipsychotic treatment in antipsychotic-naïve first-episode psychosis patients

BACKGROUND

Previous findings in psychosis have revealed mixed findings on glutamate (Glu) levels in the dorsal anterior cingulate cortex (dACC). Factors such as illness chronicity, methodology, and medication status have impeded a more nuanced evaluation of Glu in psychosis. The goal of this longitudinal neuroimaging study was to investigate the role of antipsychotics on Glu in the dACC in antipsychotic-naïve first-episode psychosis (FEP) patients.

METHODS

We enrolled 117 healthy controls (HCs) and 113 antipsychotic-naïve FEP patients for this study. 3T proton magnetic resonance spectroscopy (1H-MRS; PRESS; TE = 80 ms) data from a voxel prescribed in the dACC were collected from all participants at baseline, 6, and 16 weeks following antipsychotic treatment. Glutamate levels were quantified using the QUEST algorithm and analyzed longitudinally using linear mixed-effects models.

RESULTS

We found that baseline dACC glutamate levels in FEP were not significantly different than those of HCs. Examining Glu levels in FEP revealed a decrease in Glu levels after 16 weeks of antipsychotic treatment; this effect was weaker in HC. Finally, baseline Glu levels were associated with decreases in positive symptomology.

CONCLUSIONS

We report a progressive decrease of Glu levels over a period of 16 weeks after initiation of treatment and a baseline Glu level association with a reduction in positive symptomology, suggestive of a potential mechanism of antipsychotic drug (APD) action. Overall, these findings suggest that APDs can influence Glu within a period of 16 weeks, which has been deemed as an optimal window for symptom alleviation using APDs.

Fast High-Resolution Metabolite Mapping in the rat Brain Using 1H-FID-MRSI at 14.1 T

Magnetic resonance spectroscopic imaging (MRSI) enables the simultaneous noninvasive acquisition of MR spectra from multiple spatial locations inside the brain. Although 1H-MRSI is increasingly used in the human brain, it is not yet widely applied in the preclinical setting, mostly because of difficulties specifically related to very small nominal voxel size in the rat brain and low concentration of brain metabolites, resulting in low signal-to-noise ratio (SNR). In this context, we implemented a free induction decay 1H-MRSI sequence (1H-FID-MRSI) in the rat brain at 14.1 T. We combined the advantages of 1H-FID-MRSI with the ultra-high magnetic field to achieve higher SNR, coverage, and spatial resolution in the rat brain and developed a custom dedicated processing pipeline with a graphical user interface for Bruker 1H-FID-MRSI: MRS4Brain toolbox. LCModel fit, using the simulated metabolite basis set and in vivo measured MM, provided reliable fits for the data at acquisition delays of 1.30 ms. The resulting Cramér-Rao lower bounds were sufficiently low (< 30%) for eight metabolites of interest (total creatine, N-acetylaspartate, N-acetylaspartate + N-acetylaspartylglutamate, total choline, glutamine, glutamate, myo-inositol, and taurine), leading to highly reproducible metabolic maps. Similar spectral quality and metabolic maps were obtained with one and two averages, with slightly better contrast and brain coverage due to increased SNR in the latter case. Furthermore, the obtained metabolic maps were accurate enough to confirm the previously known brain regional distribution of some metabolites. The acquisitions proved high reproducibility over time. We demonstrated that the increased SNR and spectral resolution at 14.1 T can be translated into high spatial resolution in 1H-FID-MRSI of the rat brain in 13 min using the sequence and processing pipeline described herein. High-resolution 1H-FID-MRSI at 14.1 T provided robust, reproducible, and high-quality metabolic mapping of brain metabolites with minimal technical limitations.

Examination of methods to separate overlapping metabolites at 7T

PURPOSE

Neurochemicals of interest quantified by MRS are often composites of overlapping signals. At higher field strengths (i.e., 7T), there is better separation of these signals. As the availability of higher field strengths is increasing, it is important to re-evaluate the separability of overlapping metabolite signals.

METHODS

This study compares the ability of stimulated echo acquisition mode (STEAM-8; TE = 8 ms), short-TE semi-LASER (sLASER-34; TE = 34 ms), and long-TE semi-LASER (sLASER-105; TE = 105 ms) acquisitions to separate the commonly acquired neurochemicals at 7T (Glx, consisting of glutamate and glutamine; total N-acetyl aspartate, consisting of N-acetyl aspartate and N-acetylaspartylglutamate; total creatine, consisting of creatine and phosphocreatine; and total choline, consisting of choline, phosphocholine, and glycerophosphocholine).

RESULTS

sLASER-34 produced the lowest fit errors for most neurochemicals; however, STEAM-8 had better within-subject reproducibility and required fewer subjects to detect a change between groups. However, this is dependent on the neurochemical of interest.

CONCLUSION

We recommend short-TE STEAM for separation of most standard neurochemicals at 7T over short-TE or long-TE sLASER.

Reproducibility Made Easy: A Tool for Methodological Transparency and Efficient Standardized Reporting based on the proposed MRSinMRS Consensus

Recent expert consensus publications have highlighted the issue of poor reproducibility in magnetic resonance spectroscopy (MRS) studies, mainly due to the lack of standardized reporting criteria, which affects their clinical applicability. To combat this, guidelines for minimum reporting standards (MRSinMRS) were introduced to aid journal editors and reviewers in ensuring the comprehensive documentation of essential MRS study parameters. Despite these efforts, the implementation of MRSinMRS standards has been slow, attributed to the diverse nomenclature used by different vendors, the variety of raw MRS data formats, and the absence of appropriate software tools for identifying and reporting necessary parameters. To overcome this obstacle, we have developed the REproducibility Made Easy (REMY) standalone toolbox. REMY software supports a range of MRS data formats from major vendors like GE (p. file), Siemens (Twix, .rda, .dcm), Philips (.spar/.sdat), and Bruker (.method), and MRS-NIfTI (.nii/.json) files (.nii/.nii.gz) facilitating easy data import and export through a user-friendly interface. REMY employs external libraries such as spec2nii and pymapVBVD to accurately read and process these diverse data formats, translating complex header information into a comprehensive structure that adheres to consensus reporting standards, thereby ensuring compatibility and ease of use for researchers in generating reproducible MRS research outputs. Users can select and import datasets, choose the appropriate vendor and data format, and then generate an MRSinMRS table, log file, and methodological documents in both Latex and PDF formats. No coding knowledge is required, making the tool accessible to a wider range of users, including researchers and clinicians without programming expertise. This eliminates technical challenges related to data formatting and reporting. REMY effectively populated key sections of the MRSinMRS table with data from all supported file types. Accurate generation of hardware parameters including field strength, manufacturer, and scanner software version were demonstrated. However, it could not input data for RF coil and additional hardware information due to their absence in the files. For the acquisition section, REMY accurately read and populated fields for the pulse sequence name, nominal voxel size, repetition time (TR), echo time (TE), number of acquisitions/excitations/shots, spectral width [Hz], and number of spectral points, significantly contributing to the completion of the 'Acquisition' fields of the table. Furthermore, REMY generates a boilerplate methods text section for manuscripts.The use of REMY will facilitate more widespread adoption of the MRSinMRS checklist within the MRS community, making it easier to write and report acquisition parameters effectively.

Unraveling the pathophysiology of restless legs syndrome from multimodal MRI techniques: A systematic review

BACKGROUND

Restless Legs Syndrome (RLS) is a common neurological disorder currently diagnosed based on clinical features only, and characterized by a compulsive urge to move the legs triggered by rest or diminished arousal. This systematic review aimed at integrating all current brain magnetic resonance imaging (MRI) modalities for a convergent pathophysiological understanding of RLS phenomenology.

METHODS

We performed a MEDLINE (PubMed)-based systematic review for research articles in patients with primary RLS published in English from 2010 till November 2023. Studies meeting the inclusion criteria according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses criteria were systematically assessed for quality using modality-specific checklists, bias using AXIS tool and a narrative synthesis of the results was conducted.

RESULTS

A total of 49 studies (22 structural, 12 DTI, 7 iron-imaging, 4 spectroscopy with 10 datasets combining multiple approaches) involving 1273 patients (414 males) and 1333 healthy controls (478 males) met the eligibility criteria. Despite participant, technical/device-related and statistical heterogeneity, most agree that patients with primary RLS have structural and metabolite alterations, changes in multiple white matter tract architectures, and disrupted functional connectivity within multiple brain areas. Most of the studies (n = 43, 88 %) have a low-risk of bias on the AXIS scale. Scores on the modality-specific checklist ranged from 46 to 92 %, 70-93 % and 54-92 % for structural MRI, DTI and MRS Datasets, respectively.

CONCLUSIONS

Notwithstanding the large heterogeneity in the methods employed, global connectivity alterations suggest the utility of casting RLS within a system-level perspective rather than viewing it as related to the dysfunction of a single or particular brain region. A holistic approach and its integration within the framework of molecular vulnerability and neurotransmitter alterations are warranted to disentangle the complex pathophysiology of RLS and to identify new therapeutic targets.

Advanced Imaging in the Diagnosis and Response Assessment of High-Grade Glioma: AJR Expert Panel Narrative Review

This AJR Expert Panel Narrative Review explores the current status of advanced MRI and PET techniques for the posttherapeutic response assessment of high-grade adult-type gliomas, focusing on ongoing clinical controversies in current practice. Discussed techniques that complement conventional MRI and aid the differentiation of recurrent tumor from posttreatment effects include DWI and diffusion-tensor imaging; perfusion MRI techniques including dynamic susceptibility contrast (DSC), dynamic contrast-enhanced, and arterial spin labeling MRI; MR spectroscopy (MRS) including assessment of 2-hydroxyglutarate (2HG) concentration; glucose- and amino acid (AA)-based PET; and amide proton transfer imaging. Updated criteria for the Response Assessment in Neuro-Oncology are presented. Given the abundant supporting clinical evidence, the panel supports a recommendation that routine response assessment after high-grade glioma treatment should include perfusion MRI, particularly given the development of a consensus recommended DSC-MRI protocol. Although published studies support 2HG MRS and AA PET, these techniques' widespread adoption will likely require increased availability (for 2HG MRS) or increased insurance funding in the United States (for AA PET). The review concludes with a series of consensus opinions from the author panel, centered on the clinical integration of the advanced imaging techniques into posttreatment surveillance protocols.

A novel automated pipeline to assess MR spectroscopy quality control: Comparing current standards and manual assessment

BACKGROUND AND PURPOSE

The absence of a consensus data quality control (DQC) process inhibits the widespread adoption of MR spectroscopy. Poor DQC can lead to unreliable clinical diagnosis and irreproducible research conclusions. Currently, manual visual assessment or the standard quantitative metrics of signal-to-noise, linewidth, and model fit are used as classifiers, but these measures may not be sufficient. To supplement standard metrics, this paper proposes a novel automated DQC pipeline named Visual Evaluative Control Technology Of Resonance Spectroscopy (VECTORS).

METHODS

Manual DQC ratings were conducted on 7180 spectra obtained from 110 young adults using short-echo chemical shift imaging at 3 Tesla. Four reviewers conducted manual ratings on the presence of artifacts and location of metabolites. The ratings were labor intensive, taking over 180 hours. VECTORS was developed to quantify their DQC criteria, detecting artifacts that present as duplicate peaks, vertical shifts, and glutamine + glutamate and myoinositol peak shapes. Run on the same data using a standard laptop, VECTORS only took 2 hours.

RESULTS

The manual ratings were not monotonic to the standard quantitative metrics. VECTORS correctly flagged spectra that the manual ratings missed. VECTORS accurately flagged an additional 126 poor DQ spectra that consensus cutoffs of the standard quantitative metrics deemed good DQ.

CONCLUSION

Standard quantitative metrics may not account for all DQC artifacts as they are not monotonic to the manual ratings. However, manual ratings are labor intensive, subjective, and irreproducible. VECTORS addresses these issues and should be used in conjunction with standard quantitative metrics.

2D 1H sLASER Long-TE and 3D 31P Chemical Shift Imaging at 3 T for Monitoring Fasting-Induced Changes in Brain Tumor Tissue

BACKGROUND

Emerging evidence suggests that fasting could play a key role in cancer treatment. Its metabolic effects on gliomas require further investigation.

PURPOSE

To design a multi-voxel 1H/31P MR-spectroscopic imaging (MRSI) protocol for noninvasive metabolic monitoring of cerebral, fasting-induced changes on an individual patient/tumor level, and to assess its technical reliability/reproducibility.

STUDY TYPE

Prospective.

POPULATION

MRS phantom. Twenty-two patients (mean age = 61, 6 female) with suspected WHO grade II-IV glioma examined before and after 72-hour-fasting prior to biopsy/resection.

FIELD STRENGTH/SEQUENCE

3-T, 1H decoupled 3D 31P MRSI, 2D 1H sLASER MRSI at an echo time of 144 msec, 2D 1H MRSI (as water reference), T1-weighted, T1-weighted contrast-enhanced, T2-weighted, and FLAIR. sLASER and PRESS sequences were used for phantom measurements.

ASSESSMENT

Phantom measurements and spectral simulations were performed with various echo-times for protocol optimization. In vivo spectral analyses were conducted using LCModel and AMARES, obtaining quality/fitting parameters (linewidth, signal-to-noise-ratio, and uncertainty measures of fitting) and metabolite intensities. The volume of glioma sub-regions was calculated and correlated with MRS findings. Ex-vivo spectra of necrotic tumor tissues were obtained using high-resolution magic-angle spinning (HR-MAS) technique.

STATISTICAL TESTS

Wilcoxon signed-rank test, Bland-Altman plots, and coefficient of variation were used for repeatability analysis of quality/fitting parameters and metabolite concentrations. Spearman ρ correlation for the concentration of ketone bodies with volumes of glioma sub-regions was determined. A P-value <0.05 was considered statistically significant.

RESULTS

1H and 31P repeatability measures were highly consistent between the two sessions. β-hydroxybutyrate and acetoacetate were detectable (fitting-uncertainty <50%) in glioma sub-regions of all patients who completed the 72-hour-fasting cycle. β-hydroxybutyrate accumulation was significantly correlated with the necrotic/non-enhancing tumor core volume (ρ = 0.81) and validated using ex-vivo 1H HR-MAS.

DATA CONCLUSION

We propose a comprehensive MRS protocol that may be used for monitoring cerebral, fasting-induced changes in patients with glioma.

EVIDENCE LEVEL

1 TECHNICAL EFFICACY: Stage 4.

Optimization of 1H-MRS methods for large-volume acquisition of low-concentration downfield resonances at 3 T and 7 T

PURPOSE

This goal of this study was to optimize spectrally selective 1H-MRS methods for large-volume acquisition of low-concentration metabolites with downfield resonances at 7 T and 3 T, with particular attention paid to detection of nicotinamide adenine dinucleotide (NAD+) and tryptophan.

METHODS

Spectrally selective excitation was used to avoid magnetization-transfer effects with water, and various sinc pulses were compared with a band-selective, uniform response, pure-phase (E-BURP) pulse. Localization using a single-slice selective pulse was compared with voxel-based localization that used three orthogonal refocusing pulses, and low bandwidth refocusing pulses were used to take advantage of the chemical shift displacement of water. A technique for water sideband removal was added, and a method of coil channel combination for large volumes was introduced.

RESULTS

Proposed methods were compared qualitatively with previously reported techniques at 7 T. Sinc pulses resulted in reduced water signal excitation and improved spectral quality, with a symmetric, low bandwidth-time product pulse performing best. Single-slice localization allowed shorter TEs with large volumes, enhancing signal, whereas low-bandwidth slice-selective localization greatly reduced the observed water signal. Gradient cycling helped remove water sidebands, and frequency aligning and pruning individual channels narrowed spectral linewidths. High-quality brain spectra of NAD+ and tryptophan are shown in 4 subjects at 3 T.

CONCLUSION

Improved spectral quality with higher downfield signal, shorter TE, lower nuisance signal, reduced artifacts, and narrower peaks was realized at 7 T. These methodological improvements allowed for previously unachievable detection of NAD+ and tryptophan in human brain at 3 T in under 5 min.

On the impact of B0 shimming algorithms on single-voxel MR spectroscopy

PURPOSE

To assess the impact of different B0 shimming algorithms on MRS.

METHODS

B0 field maps and single-voxel MR spectroscopy were acquired in the prefrontal cortex of five volunteers at 3 T using five different B0 shimming approaches. B0 shimming was achieved using Siemens' proprietary shim algorithm, in addition to the Pseudo-Inverse (PI), Quadratic Programming (QuadProg), Least Squares (LSq), and Gradient optimization (Grad) algorithms. The standard deviation of the shimmed B0 field, as well as the SNR and FWHM of the measured metabolites, was used to evaluate the performance of each B0 shimming algorithm.

RESULTS

Compared to Siemens's shim, significant reductions (p < 0.01) in the standard deviation of the B0 field distribution within the MRS voxel were observed for the PI, QuadProg, and Grad algorithms (3.8 Hz, 7.3 Hz, and 3.9 Hz respectively, compared to 11.5 Hz for Siemens), but not for the LSq (12.9 Hz) algorithm. Moreover, significantly increased SNR and reduced FWHM for the N-acetylaspartate metabolite were consistent with the improvement in B0 homogeneity for the aforementioned shimming algorithms.

CONCLUSION

Here, we demonstrate that the choice of B0 shimming algorithm can have a significant impact on the quality of MR spectra and that significant improvements in spectrum quality could be achieved by using alternatives to the default vendor approach.

Detecting Tumor-Associated Intracranial Hemorrhage Using Proton Magnetic Resonance Spectroscopy

Intracranial hemorrhage associated with primary or metastatic brain tumors is a critical condition that requires urgent intervention, often through open surgery. Nevertheless, surgical interventions may not always be feasible due to two main reasons: (1) extensive hemorrhage can obscure the underlying tumor mass, limiting radiological assessment; and (2) intracranial hemorrhage may occasionally present as the first symptom of a brain tumor without prior knowledge of its existence. The current review of case studies suggests that advanced radiological imaging techniques can improve diagnostic power for tumoral hemorrhage. Adding proton magnetic resonance spectroscopy (1H-MRS), which profiles biochemical composition of mass lesions could be valuable: it provides unique information about tumor states distinct from hemorrhagic lesions bypassing the structural obliteration caused by the hemorrhage. Recent advances in 1H-MRS techniques may enhance the modality's reliability in clinical practice. This perspective proposes that 1H-MRS can be utilized in clinical settings to enhance diagnostic power in identifying tumors underlying intracranial hemorrhage.

Exploring the Effects of Prefrontal Transcranial Direct Current Stimulation on Brain Metabolites: A Concurrent tDCS-MRS Study

Transcranial Direct Current Stimulation (tDCS) is a non-invasive brain stimulation technique used to modulates cortical brain activity. However, its effects on brain metabolites within the dorsolateral prefrontal cortex (DLPFC), a crucial area targeted for brain stimulation in mental disorders, remain unclear. This study aimed to investigate whether prefrontal tDCS over the left and right DLPFC modulates levels of key metabolites, including gamma-aminobutyric acid (GABA), glutamate (Glu), glutamine/glutamate (Glx), N-acetylaspartate (NAA), near to the target region and to explore potential sex-specific effects on these metabolite concentrations. A total of 41 healthy individuals (19 female, M_age = 25 years, SD = 3.15) underwent either bifrontal active (2 mA for 20 min) or sham tDCS targeting the left (anode: F3) and right (cathode: F4) DLPFC within a 3 Tesla MRI scanner. Magnetic resonance spectroscopy (MRS) was used to monitor neurometabolic changes before, during, and after 40 min of tDCS, with measurements of two 10-min intervals during stimulation. A single voxel beneath F3 was used for metabolic quantification. Results showed a statistically significant increase in Glx levels under active tDCS compared to the sham condition, particularly during the second 10-min window and persisting into the post-stimulation phase. No significant changes were observed in other metabolites, but consistent sex differences were detected. Specifically, females showed lower levels of NAA and GABA under active tDCS compared to the sham condition, while no significant changes were observed in males. E-field modeling showed no significant differences in field magnitudes between sexes, and the magnitude of the e-fields did not correlate with changes in Glx levels between active and sham stimulation during the second interval or post-stimulation. This study demonstrates that a single session of prefrontal tDCS significantly elevates Glx levels in the left DLPFC, with effects persisting post-stimulation. However, the observed sex differences in the neurochemical response to tDCS were not linked to specific stimulation intervals or variations in e-field magnitudes, highlighting the complexity of tDCS effects and the need for personalized neuromodulation strategies.

A study of long-term GABA and high-energy phosphate alterations in the primary motor cortex using anodal tDCS and 1H/31P MR spectroscopy

Introduction

Anodal transcranial direct current stimulation (tDCS) has been reported to modulate gamma-aminobutyric acid levels and cerebral energy consumption in the brain. This study aims to investigate long-term GABA and cerebral energy modulation following anodal tDCS over the primary motor cortex.

Method

To assess GABA and energy level changes, proton and phosphorus magnetic resonance spectroscopy data were acquired before and after anodal or sham tDCS. In anodal stimulation, a 1 mA current was applied for 20 min, and the duration of ramping the current up/down at the start and end of the intervention was 10 s. In the sham-stimulation condition, the current was first ramped up over a period of 10 s, then immediately ramped down, and the condition was maintained for the next 20 min.

Results

The GABA concentration increased significantly following anodal stimulation in the first and second post-stimulation measurements. Likewise, both ATP/Pi and PCr/Pi ratios increased after anodal stimulation in the first and second post-stimulation measurements.

Conclusion

The approach employed in this study shows the feasibility of measuring long-term modulation of GABA and high-energy phosphates following anodal tDCS targeting the left M1, offering valuable insights into the mechanisms of neuroplasticity and energy metabolism, which may have implications for applications of this intervention in clinical populations.

Improving Brain Metabolite Detection with a Combined Low-Rank Approximation and Denoising Diffusion Probabilistic Model Approach

In vivo proton magnetic resonance spectroscopy (MRS) is a noninvasive technique for monitoring brain metabolites. However, it is challenged by a low signal-to-noise ratio (SNR), often necessitating extended scan times to compensate. One of the conventional techniques for noise reduction is signal averaging, which is inherently time-consuming and can lead to participant discomfort, thus posing limitations in clinical settings. This study aimed to develop a hybrid denoising strategy that integrates low-rank approximation and denoising diffusion probabilistic model (DDPM) to enhance MRS data quality and shorten scan times. Using publicly available 1H MRS datasets from 15 subjects, we applied the Casorati SVD and DDPM to obtain baseline and functional data during a pain stimulation task. This method significantly improved SNR, resulting in outcomes comparable to or better than averaging over 32 signals. It also provided the most consistent metabolite measurements and adequately tracked temporal changes in glutamate levels, correlating with pain intensity ratings after heating. These findings demonstrate that our approach enhances MRS data quality, offering a more efficient alternative to conventional methods and expanding the potential for the real-time monitoring of neurochemical changes. This contribution has the potential to advance MRS techniques by integrating advanced denoising methods to increase the acquisition speed and enhance the precision of brain metabolite analyses.

Metabolic Profile of Cerebellum in Posterior Fossa Tumor Survivors: Correlation With Memory Impairment

AIMS

The cerebellum is a key structure in working and procedural memory. The aim of the present prospective exploratory study was to investigate, the metabolic characteristics of the cerebellum in posterior fossa tumor (PFT) survivors using 3D proton magnetic resonance spectroscopy imaging (3D MRSI), to determine whether metabolites could be useful biomarkers of memory impairment.

MATERIALS AND METHODS

Sixty participants were included in the IMPALA study, divided into three groups: 22 irradiated PFT, 17 nonirradiated PFT, and 21 healthy controls matched with irradiated PFT for age, sex, and handedness. PFT survivors were treated at least 5 years ago, either by surgery or a combination of surgery, chemotherapy, and radiotherapy. All participants underwent working and procedural memory tests and multimodal MRI including a 3D MRSI sequence. N-acetylaspartate (NAA), choline (Cho), creatine (Cr), and lactate (Lac) metabolite values were extracted from the cerebellum for comparisons between groups, correlations with neurocognitive test scores, and radiotherapy doses.

RESULTS

Median (range) age at neurocognitive tests was 18 (7-26) years. Median Cho, Cr, NAA, and Lac values, and the ratio of NAA to the sum of metabolites were significantly lower for PFT survivors than for healthy controls (p < 0.05). Scores on working and procedural memory tests were significantly lower for PFT survivors (p < 0.004) and correlated with median and maximum Cho and NAA values (0.28

CONCLUSION

Results revealed changes in cerebellar metabolic values in PFT survivors that were closely correlated with memory deficits, suggesting that some metabolites could be used as markers of cognitive decline, but this will require validation on a larger sample size.

Collapse

Key Words

• Cerebellum
• Memory
• Metabolic
• Paediatric posterior fossa tumor
• Radiotherapy
• Spectroscopy imaging

Collapse

MESH Headings

• Humans
• Female
• Male
• Cerebellum/metabolism
• Cerebellum/radiation effects
• Cerebellum/diagnostic imaging
• Adult
• Infratentorial Neoplasms/metabolism
• Prospective Studies
• Memory Disorders/etiology
• Memory Disorders/metabolism
• Cancer Survivors
• Young Adult
• Adolescent
• Case-Control Studies
• Metabolome
• Choline/metabolism

Collapse

Grants Collapse

Affiliation(s)

F Tensaouti Radiation Oncology Department, Oncopole Claudius Regaud- Institut Universitaire du Cancer de Toulouse-Oncopole, Toulouse, France; ToNIC, Toulouse NeuroImaging Center, Université de Toulouse, Inserm, UPS, France.
N Courbière ToNIC, Toulouse NeuroImaging Center, Université de Toulouse, Inserm, UPS, France
B Cabarrou Biostatistics & Health Data Science Unit, Oncopole Claudius Regaud/Institut Universitaire du Cancer de Toulouse - Oncopôle, Toulouse, France
L Pollidoro ToNIC, Toulouse NeuroImaging Center, Université de Toulouse, Inserm, UPS, France
M Roques Radiology Department, Toulouse University Hospital, Toulouse, France
A Sévely Radiology Department, Toulouse University Hospital, Toulouse, France
P Péran ToNIC, Toulouse NeuroImaging Center, Université de Toulouse, Inserm, UPS, France
E Baudou ToNIC, Toulouse NeuroImaging Center, Université de Toulouse, Inserm, UPS, France; Pediatric Neurology Department, Children's Hospital, Toulouse University Hospital, Toulouse, France
A Laprie Radiation Oncology Department, Oncopole Claudius Regaud- Institut Universitaire du Cancer de Toulouse-Oncopole, Toulouse, France; ToNIC, Toulouse NeuroImaging Center, Université de Toulouse, Inserm, UPS, France

Collapse

Collaborators Collapse

Regional variations in cingulate cortex glutamate levels: a magnetic resonance spectroscopy study at 3 T

Regional variations in glutamate levels across the cingulate cortex, decreasing rostral to caudal, have been observed previously in healthy volunteers with proton magnetic resonance spectroscopy (1H-MRS) at 7 T. This study sought to explore cingulate cortex glutamate trends further by investigating whether a similar gradient could be detected at 3 T, the effect of sex, as well as whether individual variations gave rise to more than one regional glutamate pattern. 1H-MRS at 3 T [Phillips Elition; semi-localization by adiabatic selective refocusing, echo time (TE)/repetition time (TR) = 32/5,000] was acquired in four cingulate regions: the anterior, midanterior, midposterior, and posterior cortices, in 50 healthy participants (26 F) scanned at a fixed time of day and with controlled food intake. K-means clustering was used to characterize the presence of distinct regional patterns, which were then compared between sex and clusters. In addition, cortical thickness was compared between clusters and in relation to glutamate. Aligned with 7 T findings, we demonstrated that average glutamate levels decreased rostral to caudal in the healthy cingulate cortex. No effect of sex was found, suggesting similar resting glutamate levels in both sexes. Interestingly, the majority of participants were characterized by glutamate levels that did not significantly change across the cingulate (65%). Different regional patterns in cortical thickness between clusters offer further evidence into these distinct glutamate variations and suggest that both a neuroanatomical and a functional role may lead to these findings. This study provides a much-needed foundation for further research to determine the implications of neurotransmission patterns in health and disease.NEW & NOTEWORTHY In a large, sex-balanced sample of healthy individuals, we demonstrate that average regional differences (rostral to caudal) in cingulate cortex glutamate exist, using optimized experimental conditions and 3 T magnetic resonance spectroscopy techniques. Results align with observations from 7 T. A novel clustering approach was introduced to determine the number of patterns for glutamate in the healthy adult brain for the first time. These findings demonstrate that regional differences are detectable at 3 T when present and suggest the occurrence of multiple glutamate metabolism patterns in the cingulate.

Reporting checklists in neuroimaging: promoting transparency, replicability, and reproducibility

Neuroimaging plays a crucial role in understanding brain structure and function, but the lack of transparency, reproducibility, and reliability of findings is a significant obstacle for the field. To address these challenges, there are ongoing efforts to develop reporting checklists for neuroimaging studies to improve the reporting of fundamental aspects of study design and execution. In this review, we first define what we mean by a neuroimaging reporting checklist and then discuss how a reporting checklist can be developed and implemented. We consider the core values that should inform checklist design, including transparency, repeatability, data sharing, diversity, and supporting innovations. We then share experiences with currently available neuroimaging checklists. We review the motivation for creating checklists and whether checklists achieve their intended objectives, before proposing a development cycle for neuroimaging reporting checklists and describing each implementation step. We emphasize the importance of reporting checklists in enhancing the quality of data repositories and consortia, how they can support education and best practices, and how emerging computational methods, like artificial intelligence, can help checklist development and adherence. We also highlight the role that funding agencies and global collaborations can play in supporting the adoption of neuroimaging reporting checklists. We hope this review will encourage better adherence to available checklists and promote the development of new ones, and ultimately increase the quality, transparency, and reproducibility of neuroimaging research.

Diffusion magnetic resonance spectroscopy captures microglial reactivity related to gut-derived systemic lipopolysaccharide: A preliminary study

Neuroinflammation is a key component underlying multiple neurological disorders, yet non-invasive and cost-effective assessment of in vivo neuroinflammatory processes in the central nervous system remains challenging. Diffusion weighted magnetic resonance spectroscopy (dMRS) has shown promise in addressing these challenges by measuring diffusivity properties of different neurometabolites, which can reflect cell-specific morphologies. Prior work has demonstrated dMRS utility in capturing microglial reactivity in the context of lipopolysaccharide (LPS) challenges and serious neurological disorders, detected as changes of microglial metabolite diffusivity properties. However, the extent to which such dMRS metrics are capable of detecting subtler and more nuanced levels of neuroinflammation in populations without overt neuropathology is unknown. Here we examined the relationship between intrinsic, gut-derived levels of systemic LPS and dMRS-based apparent diffusion coefficients (ADC) of choline, creatine, and N-acetylaspartate (NAA) in two brain regions: the thalamus and the corona radiata. Higher plasma LPS concentrations were significantly associated with increased ADC of choline and NAA in the thalamic region, with no such relationships observed in the corona radiata for any of the metabolites examined. As such, dMRS may have the sensitivity to measure microglial reactivity across populations with highly variable levels of neuroinflammation, and holds promising potential for widespread applications in both research and clinical settings.

Model-based frequency-and-phase correction of 1H MRS data with 2D linear-combination modeling

PURPOSE

Retrospective frequency-and-phase correction (FPC) methods attempt to remove frequency-and-phase variations between transients to improve the quality of the averaged MR spectrum. However, traditional FPC methods like spectral registration struggle at low SNR. Here, we propose a method that directly integrates FPC into a 2D linear-combination model (2D-LCM) of individual transients ("model-based FPC"). We investigated how model-based FPC performs compared to the traditional approach, i.e., spectral registration followed by 1D-LCM in estimating frequency-and-phase drifts and, consequentially, metabolite level estimates.

METHODS

We created synthetic in-vivo-like 64-transient short-TE sLASER datasets with 100 noise realizations at 5 SNR levels and added randomly sampled frequency and phase variations. We then used this synthetic dataset to compare the performance of 2D-LCM with the traditional approach (spectral registration, averaging, then 1D-LCM). Outcome measures were the frequency/phase/amplitude errors, the SD of those ground-truth errors, and amplitude Cramér Rao lower bounds (CRLBs). We further tested the proposed method on publicly available in-vivo short-TE PRESS data.

RESULTS

2D-LCM estimates (and accounts for) frequency-and-phase variations directly from uncorrected data with equivalent or better fidelity than the conventional approach. Furthermore, 2D-LCM metabolite amplitude estimates were at least as accurate, precise, and certain as the conventionally derived estimates. 2D-LCM estimation of FPC and amplitudes performed substantially better at low-to-very-low SNR.

CONCLUSION

Model-based FPC with 2D linear-combination modeling is feasible and has great potential to improve metabolite level estimation for conventional and dynamic MRS data, especially for low-SNR conditions, for example, long TEs or strong diffusion weighting.

Heterogeneity and convergence across seven neuroimaging modalities: a review of the autism spectrum disorder literature

Background

A growing body of literature classifies autism spectrum disorder (ASD) as a heterogeneous, complex neurodevelopmental disorder that often is identified prior to three years of age. We aim to provide a narrative review of key structural and functional properties that differentiate the neuroimaging profile of autistic youth from their typically developing (TD) peers across different neuroimaging modalities.

Methods

Relevant studies were identified by searching for key terms in PubMed, with the most recent search conducted on September 1, 2023. Original research papers were included if they applied at least one of seven neuroimaging modalities (structural MRI, functional MRI, DTI, MRS, fNIRS, MEG, EEG) to compare autistic children or those with a family history of ASD to TD youth or those without ASD family history; included only participants <18 years; and were published from 2013 to 2023.

Results

In total, 172 papers were considered for qualitative synthesis. When comparing ASD to TD groups, structural MRI-based papers (n = 26) indicated larger subcortical gray matter volume in ASD groups. DTI-based papers (n = 14) reported higher mean and radial diffusivity in ASD participants. Functional MRI-based papers (n = 41) reported a substantial number of between-network functional connectivity findings in both directions. MRS-based papers (n = 19) demonstrated higher metabolite markers of excitatory neurotransmission and lower inhibitory markers in ASD groups. fNIRS-based papers (n = 20) reported lower oxygenated hemoglobin signals in ASD. Converging findings in MEG- (n = 20) and EEG-based (n = 32) papers indicated lower event-related potential and field amplitudes in ASD groups. Findings in the anterior cingulate cortex, insula, prefrontal cortex, amygdala, thalamus, cerebellum, corpus callosum, and default mode network appeared numerous times across modalities and provided opportunities for multimodal qualitative analysis.

Conclusions

Comparing across neuroimaging modalities, we found significant differences between the ASD and TD neuroimaging profile in addition to substantial heterogeneity. Inconsistent results are frequently seen within imaging modalities, comparable study populations and research designs. Still, converging patterns across imaging modalities support various existing theories on ASD.

High-quality lipid suppression and B0 shimming for human brain 1H MRSI

Magnetic Resonance Spectroscopic Imaging (MRSI) is a powerful technique that can map the metabolic profile in the brain non-invasively. Extracranial lipid contamination and insufficient B0 homogeneity however hampers robustness, and as a result has hindered widespread use of MRSI in clinical and research settings. Over the last six years we have developed highly effective extracranial lipid suppression methods with a second order gradient insert (ECLIPSE) utilizing inner volume selection (IVS) and outer volume suppression (OVS) methods. While ECLIPSE provides > 100-fold in lipid suppression with modest radio frequency (RF) power requirements and immunity to B1+ field variations, axial coverage is reduced for non-elliptical head shapes. In this work we detail the design, construction, and utility of MC-ECLIPSE, a pulsed second order gradient coil with Z2 and X2Y2 fields, combined with a 54-channel multi-coil (MC) array. The MC-ECLIPSE platform allows arbitrary region of interest (ROI) shaped OVS for full-axial slice coverage, in addition to MC-based B0 field shimming, for robust human brain proton MRSI. In vivo experiments demonstrate that MC-ECLIPSE allows axial brain coverage of 92-95 % is achieved following arbitrary ROI shaped OVS for various head shapes. The standard deviation (SD) of the residual B0 field following SH2 and MC shimming were 25 ± 9 Hz and 18 ± 8 Hz over a 5 cm slab, and 18 ± 5 Hz and 14 ± 6 Hz over a 1.5 cm slab, respectively. These results demonstrate that B0 magnetic field shimming with the MC array supersedes second order harmonic capabilities available on standard MRI systems for both restricted and large ROIs. Furthermore, MC based B0 shimming provides comparable shimming performance to an unrestricted SH5 shim set for both restricted, and 5-cm slab shim challenges. Phantom experiments demonstrate the high level of localization performance achievable with MC-ECLIPSE, with ROI edge chemical shift displacements ranging from 1-3 mm with a median value of 2 mm, and transition width metrics ranging from 1-2.5 mm throughout the ROI edge. Furthermore, MC based B0 shimming is comparable to performance following a full set of unrestricted spherical harmonic fields up to order 5. Short echo time MRSI and GABA-edited MRSI acquisitions in the human brain following MC-shimming and arbitrary ROI shaping demonstrate full-axial slice coverage and extracranial lipid artifact free spectra. MC-ECLIPSE allows full-axial coverage and robust MRSI acquisitions, while allowing interrogation of cortical tissue proximal to the skull, which has significant value in a wide range of neurological and psychiatric conditions.

GABA-edited MEGA-PRESS at 3 T: Does a measured macromolecule background improve linear combination modeling?

PURPOSE

The J-difference edited γ-aminobutyric acid (GABA) signal is contaminated by other co-edited signals-the largest of which originates from co-edited macromolecules (MMs)-and is consequently often reported as "GABA+." MM signals are broader and less well-characterized than the metabolites, and are commonly approximated using a Gaussian model parameterization. Experimentally measured MM signals are a consensus-recommended alternative to parameterized modeling; however, they are relatively under-studied in the context of edited MRS.

METHODS

To address this limitation in the literature, we have acquired GABA-edited MEGA-PRESS data with pre-inversion to null metabolite signals in 13 healthy controls. An experimental MM basis function was derived from the mean across subjects. We further derived a new parameterization of the MM signals from the experimental data, using multiple Gaussians to accurately represent their observed asymmetry. The previous single-Gaussian parameterization, mean experimental MM spectrum and new multi-Gaussian parameterization were compared in a three-way analysis of a public MEGA-PRESS dataset of 61 healthy participants.

RESULTS

Both the experimental MMs and the multi-Gaussian parameterization exhibited reduced fit residuals compared to the single-Gaussian approach (p = 0.034 and p = 0.031, respectively), suggesting they better represent the underlying data than the single-Gaussian parameterization. Furthermore, both experimentally derived models estimated larger MM fractional contribution to the GABA+ signal for the experimental MMs (58%) and multi-Gaussian parameterization (58%), compared to the single-Gaussian approach (50%).

CONCLUSIONS

Our results indicate that single-Gaussian parameterization of edited MM signals is insufficient and that both experimentally derived GABA+ spectra and their parameterized replicas improve the modeling of GABA+ spectra.