Limb-girdle muscular dystrophy R2-dysferlin related (LGMD-R2) is a progressive muscle condition with marked variability in disease course, making prognosis challenging. Quantitative MRI (qMRI) has emerged as a complementary tool that may detect progression earlier and more precisely. Integrating different data modalities is challenging with conventional approaches, and artificial intelligence (AI) can help overcome this. Our aim is to develop robust models capable of predicting clinical progression in LGMD-R2 by incorporating AI-based techniques into the analysis pipeline. Data from 188 COS 1 participants were analysed. Disease progression was assessed using the North Star Assessment for Limb Girdle type Muscular Dystrophies (NSAD). Ambulatory individuals with a maximum NSAD ≥ 20 were included, and progression trajectories were identified through hierarchical clustering. Feature selection was performed using a machine learning pipeline, and top predictors were entered into stepwise logistic regression to build clinical-only and combined clinical-MRI models. Two stages of progression were identified, a fast one with a mean three-year loss of 14.4 NSAD points, and a moderate one, with a mean loss of 3.8 NSAD points. The combined model achieved better balanced accuracy than the clinical-only one (83.7

Background and ObjectivesLimb-girdle muscular dystrophy R2 (LGMDR2) is characterized by progressive muscle weakness usually leading to severe disability. The rate of progression and disease severity is variable among patients, although factors influencing this variability are not completely understood. The Dysferlinopathy Clinical Outcome Study is a natural history study that followed patients with LGMDR2 for 3 consecutive years using functional outcome measures and skeletal muscle MRI.The aim of our study was to develop statistical models able to describe fat fraction (FF) progression of the lower limbs in patients with LGMDR2 using clinical and radiologic variables to better understand which factors influence disease progression and improve the design of future clinical trials.Background and ObjectivesLimb-girdle muscular dystrophy R2 (LGMDR2) is characterized by progressive muscle weakness usually leading to severe disability. The rate of progression and disease severity is variable among patients, although factors influencing this variability are not completely understood. The Dysferlinopathy Clinical Outcome Study is a natural history study that followed patients with LGMDR2 for 3 consecutive years using functional outcome measures and skeletal muscle MRI.The aim of our study was to develop statistical models able to describe fat fraction (FF) progression of the lower limbs in patients with LGMDR2 using clinical and radiologic variables to better understand which factors influence disease progression and improve the design of future clinical trials.MethodsWe used linear-mixed modeling to analyze changes in FF over time according to patients' age. We calculated the average FF trajectory for each muscle of the lower limbs. We built 2 multivariate models for each segment adding other clinical factors and using likelihood ratio test and residuals' analysis to determine whether they better fitted observed FF values.ResultsMuscles that participated in the same joint movement progressed similarly over time. FF was expected to be higher the older patients were and the earlier the age at symptom onset. Women had absolute FF values 8.8% higher than men in the lower leg. No differences in FF trajectory were seen based on ethnic groups (White, Asian, Black, or Hispanic), genetic variants, or residual dysferlin expression. Although multivariate models showed a better global fit to the data, there was no improvement in representing individual patient variability.DiscussionIn conclusion, this study provides a better understanding of skeletal muscle fat replacement progression in the lower limb muscles of patients with LGMDR2, highlighting the influence of age at symptom onset, sex, and baseline motor function, which should be considered in the design and analysis of clinical trials. Although complex models improved the overall data fit, they did not improve the accuracy in identifying changes at a patient level, underlying the need for further research and validation and the fact that other variables we have not measured are probably influencing progression.

Muscle twitch dynamics and fatigability change in response to muscle disease. In this study, we developed an imaging paradigm to measure muscle twitch dynamics, and the response of the muscle to voluntary fatiguing contractions. We used a novel imaging technique called motor unit magnetic resonance imaging (MUMRI). MUMRI allows visualisation of muscle and motor unit activity by combining in-scanner electrical stimulation with dynamic pulsed gradient spin echo (twitch dynamics, PGSE-MUMRI) and phase contrast (fatigue, PC-MUMRI) imaging. In Part I of this study, we scanned 10 healthy controls, we measured the muscle rise (Trise), contraction (Tcontract) and half-relaxation time (Thalf-relax) of the tibialis anterior (TA) muscle on a voxel-wise basis using PGSE-MUMRI. Five controls were scanned twice to assess reproducibility; PGSE-MUMRI demonstrated reproducible results, with low variation between scans 3.4% for Trise, 6.4% for Tcontract and 7.1% for Thalf-relax. We then developed a PC-MUMRI paradigm to measure the recovery of the TA in response to a fatiguing voluntary exercise. In Part II of the study, we applied these two novel imaging paradigms in a cohort study of nine patients with single large-scale mtDNA deletion primary mitochondrial myopathy (PMM). Patients underwent a 12-week resistance exercise programme and baseline, and follow-up MRI was performed. PGSE-MUMRI detected a significantly longer muscle contraction time between baseline and follow-up in PMM patients 108.7 ± 7.9 vs. post-119.3 ± 10.4 ms; p = 0.018. There was no statistical difference in the recovery half maximum measured using PC-MUMRI in PMM patients between baseline and follow-up 254 ± 109 vs. 137 ± 41 s; p = 0.074. In conclusion, PGSE-MUMRI has detected differences in muscle twitch dynamics between controls and PMM following an exercise programme, and we can visualise differences in twitch dynamics subregions of muscle using this technique. The PC-MUMRI technique has shown promise as a novel measure of muscle fatigue.

Background and Objectives:Limb-girdle muscular dystrophy R2 (LGMDR2) is characterized by progressive muscle weakness usually leading to severe disability. The rate of progression and disease severity is variable among patients, although factors influencing this variability are not completely understood. The Dysferlinopathy Clinical Outcome Study is a natural history study that followed patients with LGMDR2 for 3 consecutive years using functional outcome measures and skeletal muscle MRI.The aim of our study was to develop statistical models able to describe fat fraction (FF) progression of the lower limbs in patients with LGMDR2 using clinical and radiologic variables to better understand which factors influence disease progression and improve the design of future clinical trials. Methods:We used linear-mixed modeling to analyze changes in FF over time according to patients' age. We calculated the average FF trajectory for each muscle of the lower limbs. We built 2 multivariate models for each segment adding other clinical factors and using likelihood ratio test and residuals' analysis to determine whether they better fitted observed FF values. Results:Muscles that participated in the same joint movement progressed similarly over time. FF was expected to be higher the older patients were and the earlier the age at symptom onset. Women had absolute FF values 8.8% higher than men in the lower leg. No differences in FF trajectory were seen based on ethnic groups (White, Asian, Black, or Hispanic), genetic variants, or residual dysferlin expression. Although multivariate models showed a better global fit to the data, there was no improvement in representing individual patient variability. Discussion:In conclusion, this study provides a better understanding of skeletal muscle fat replacement progression in the lower limb muscles of patients with LGMDR2, highlighting the influence of age at symptom onset, sex, and baseline motor function, which should be considered in the design and analysis of clinical trials. Although complex models improved the overall data fit, they did not improve the accuracy in identifying changes at a patient level, underlying the need for further research and validation and the fact that other variables we have not measured are probably influencing progression.

ABSTRACT Background Dysferlin plays a key role in cell membrane repair; its absence or malfunction in patients with dysferlin‐deficient limb girdle muscular dystrophy leads to muscle fibre death. Muscle magnetic resonance (MR) imaging allows non‐invasive and repeatable measurements that can report on pathological changes observed in dysferlinopathy patients (DP). We aimed to demonstrate the feasibility of utilising volume‐localised 23Na spectroscopy as a novel approach to characterise muscle Na+ content and biexponential T2* at rest, and dynamically post‐exercise, in patients with dysferlinopathy and in matched healthy controls. Methods Adult DP and age and sex matched healthy volunteers (HV) were recruited and scanned on a 3 T clinical MR scanner. Following baseline scans, participants performed physiotherapist‐guided isometric dorsiflexion contractions until tibialis anterior (TA) muscle exhaustion. Dynamic volume‐localised sodium‐23 (23Na)‐ and proton (1H)‐MR scans were acquired serially for 35 min post‐exercise. MR data were analysed to determine TA lipid content, change in TA sodium content, biexponential sodium T2* properties and TA water 1H T2. Results Ten DP (mean age ± standard deviation [SD]: 38.0 ± 10.8 years; 80% female) and 10 HV (mean age ± SD: 38.9 ± 11.5 years) were scanned. Baseline muscle water 1H T2 and sodium concentration were significantly higher in DP compared to matched controls (1H T2 DP [SD] = 33.8 [2.7] ms, 1H T2 HV = 29.3 [1.1] ms, p < 0.001; [23Na]DP = 36.2 [11.4] mM, [23Na]HV = 19.6 [3.1] mM, p < 0.001). 1H T2 and sodium content in healthy controls showed significant post‐exercise elevation with a slower time‐to‐peak for sodium content compared to 1H T2. 1H T2 and sodium content change post‐exercise was highly variable in the DP group. Notably, 23Na dynamics in one DP with normal muscle fat fraction were similar to HV. Biexponential 23Na T2* was measured at baseline in HV (T2*slow = 13.4 [2.3] ms, T2*fast = 2.2 [1.3] ms), and DP (T2*slow = 14.0 [1.5] ms and T2*fast = 1.0 [0.5] m). Equivalent measurements post‐exercise revealed an increase in the fraction of the slow‐relaxing component in HV (p < 0.05), consistent with oedematous changes. Conclusions Assessment of TA muscle fat fraction, 1H T2, sodium content and sodium T2* relaxation properties revealed differences at baseline and in post‐exercise dynamics between patients with dysferlinopathy and matched controls. Post‐exercise 23Na recovery dynamics followed a well‐defined time course in HV. Heterogeneous alterations in sodium content and MR relaxation properties in DP may reflect altered ion homeostasis associated with chronic muscle damage.

INTRODUCTION/AIMS:Surviving motor units in neurogenic diseases demonstrate collateral reinnervation. Scanning electromyography (EMG) reveals normal motor unit corridor length, but with "silent regions," suggesting that reinnervation does not result in increased motor unit size but may increase motor unit complexity. Motor unit magnetic resonance imaging (MUMRI) pairs MR imaging with electrical nerve stimulation to visualize individual motor units. This study aimed to assess the motor unit dimensions and complexity in patients with previous poliomyelitis compared to healthy controls. METHODS:Patients with a history of polio were recruited from the British Polio Fellowship, compared to a retrospective cohort of healthy controls. They underwent medical history and examination of lower limb power, fatigue assessment (fatigue severity score, FSS), and a 3 T MUMRI scan of the less-affected lower limb. The cross-sectional area, maximum, and minimum Feret diameter of the motor unit territories in tibialis anterior were calculated. Motor unit complexity was computed using the Hausdorff box-counting method. RESULTS:Of 12 polio survivors, n = 8 (6 female) were suitable for analysis and were compared to 19 controls. The mean motor unit maximum Feret diameter was 10.3 ± 3.1 mm compared to 8.4 ± 5.2 mm in controls (p = 0.34). The mean shape complexity was 0.59 ± 0.12 compared to 0.45 ± 0.2 in controls (p = 0.03). DISCUSSION:Polio survivors demonstrate motor units with normal dimensions but increased shape complexity, indicating nonuniform collateral reinnervation largely limited to existing territories. The size and shape of motor units could help in understanding the physiological processes behind reinnervation, both in polio and other neurogenic diseases such as amyotrophic lateral sclerosis.

Gliomas are the most common type of malignant brain tumour and are frequently associated with seizures. Seizures associated with gliomas are frequently refractory to conventional antiseizure medications and constitute a significant cause of morbidity. There is a clear need for a biomarker that will guide resection of the ictogenic areas during neurosurgical procedures to improve seizure control, limit morbidity and achieve better tumour clearance. Several studies have investigated glutamate in the peri-tumoural region using magnetic resonance spectroscopy (MRS), but findings remain inconclusive regarding the association among elevated glutamate, tumour type and seizure history. Furthermore, it is not known if high cortical glutamate would lead to spontaneous interictal discharges (IIDs) ex vivo. In this prospective cross-sectional observational study of patients undergoing surgery for supratentorial gliomas, pre-operative 1D MRS imaging using an echo time (TE) averaged Point Resolved Spectroscopy (PRESS) sequence was performed to quantify glutamate levels in the peri-tumoural region. Ex vivo human cortical extracellular local field potential (LFP) recordings from the cortical tissue were used to record the presence or absence of spontaneous IIDs. Metabolite data were compared between cortical regions with spontaneous IIDs (n = 6 patients), versus regions with no evidence of spontaneous IIDs (n = 14 patients). Our findings provide new metabolic evidence that peri-tumoural glutamate accumulation is strongly linked to spontaneous IIDs (P = 0.006), reinforcing its role in glioma-associated seizures. The lack of significant differences in choline to creatine and N-acetyl aspartate to creatine ratios between IID and non-IID regions suggests seizure susceptibility is independent of tumour infiltration or neuronal degeneration. These results highlight MRS as a promising non-invasive tool for identifying metabolically active, seizure-prone cortical regions, which could aid in refining surgical planning. However, further validation is required to determine its utility in predicting clinical seizure occurrence and guiding post-operative seizure management.

Background: People with bipolar disorder (BD) tend to show widespread cognitive impairment compared to healthy controls. Impairments in processing speed (PS), attention, and executive function (EF) may represent 'core' impairments that have a role in wider cognitive dysfunction. Cognitive impairments appear to relate to structural brain abnormalities in BD, but whether core deficits are related to particular brain regions is unclear and much of the research on brain-cognition associations is limited by univariate analysis and small samples. Methods: Euthymic BD patients (n=56) and matched healthy controls (n=26) underwent T1-weighted MRI scans and completed neuropsychological tests of PS, attention, and EF. We utilised public datasets to develop a normative model of cortical thickness (n=5,977) to generate robust estimations of cortical abnormalities in patients. Canonical correlation analysis was used to assess multivariate brain-cognition associations in BD, controlling for age, sex, and premorbid IQ. Results: BD showed impairments on tests of PS, attention, and EF, and abnormal cortical thickness in several brain regions compared to healthy controls. Impairments in tests of PS and EF were most strongly associated with cortical thickness in left inferior temporal, right entorhinal, and right temporal pole areas. Conclusion: Impairments in PS, attention, and EF can be observed in euthymic BD and may be related to abnormal cortical thickness in temporal regions. Future research should continue to leverage multivariate methods to examine complex brain-cognition associations in BD. Future research may benefit from exploring covariance between traditional brain structural morphological metrics such as cortical thickness, cortical volume, and surface area.

Diffusion-weighted imaging (DWI) is an established MRI technique that can investigate tissue microstructure at the scale of a few micrometers. Musculoskeletal tissues typically have a highly ordered structure to fulfill their functions and therefore represent an optimal application of DWI. Even more since disruption of tissue organization affects its biomechanical properties and may indicate irreversible damage. The application of DWI to the musculoskeletal system faces application-specific challenges on data acquisition including susceptibility effects, the low T2 relaxation time of most musculoskeletal tissues (2-70 msec) and the need for sub-millimetric resolution. Thus, musculoskeletal applications have been an area of development of new DWI methods. In this review, we provide an overview of the technical aspects of DWI acquisition including diffusion-weighting, MRI pulse sequences and different diffusion regimes to study tissue microstructure. For each tissue type (growth plate, articular cartilage, muscle, bone marrow, intervertebral discs, ligaments, tendons, menisci, and synovium), the rationale for the use of DWI and clinical studies in support of its use as a biomarker are presented. The review describes studies showing that DTI of the growth plate has predictive value for child growth and that DTI of articular cartilage has potential to predict the radiographic progression of joint damage in early stages of osteoarthritis. DTI has been used extensively in skeletal muscle where it has shown potential to detect microstructural and functional changes in a wide range of muscle pathologies. DWI of bone marrow showed to be a valuable tool for the diagnosis of benign and malignant acute vertebral fractures and bone metastases. DTI and diffusion kurtosis have been investigated as markers of early intervertebral disc degeneration and lower back pain. Finally, promising new applications of DTI to anterior cruciate ligament grafts and synovium are presented. The review ends with an overview of the use of DWI in clinical routine. EVIDENCE LEVEL: 5 TECHNICAL EFFICACY: Stage 3.

Magnetic resonance imaging (MRI) is routinely used in the musculoskeletal system to measure skeletal muscle structure and pathology in health and disease. Recently, it has been shown that MRI also has promise for detecting the functional changes, which occur in muscles, commonly associated with a range of neuromuscular disorders. This review focuses on novel adaptations of MRI, which can detect the activity of the functional sub-units of skeletal muscle, the motor units, referred to as "motor unit MRI (MUMRI)." MUMRI utilizes pulsed gradient spin echo, pulsed gradient stimulated echo and phase contrast MRI sequences and has, so far, been used to investigate spontaneous motor unit activity (fasciculation) and used in combination with electrical nerve stimulation to study motor unit morphology and muscle twitch dynamics. Through detection of disease driven changes in motor unit activity, MUMRI shows promise as a tool to aid in both earlier diagnosis of neuromuscular disorders and to help in furthering our understanding of the underlying mechanisms, which proceed gross structural and anatomical changes within diseased muscle. Here, we summarize evidence for the use of MUMRI in neuromuscular disorders and discuss what future research is required to translate MUMRI toward clinical practice. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY: Stage 3.

Skeletal muscle quantitative muscle magnetic resonance imaging (MRI) is sensitive to changes in fat content over time. The Dysferlinopathy Clinical Outcome Study 1 is an international, multicentre study in which patients were followed up clinically and radiologically for 3 years. Dixon scans of the lower limbs were obtained yearly for 109 patients. The aim of this study is to describe changes in fat fraction (FF) over time, correlate these with clinical characteristics and build a model that can predict FF progression over time. We used linear mixed effect modelling to analyse changes in FF over time according to patients’ age. We calculated the average FF trajectory for each muscle, for lower leg and, for thigh. We then built complex models for lower leg and thigh FF progression adding other clinical and radiological variables and using likelihood ratio test and residuals’ analysis to determine if they better predicted FF progression in this cohort. Before the age of 30, lower leg muscles showed higher FF values than thigh muscles. Thigh muscles (except for sartorius and gracilis) progressed significantly faster than lower leg muscles (ANOVA, p<0.05). The more severely affected muscles at age 40 (median age of our cohort) were the hamstrings, adductor magnus and gastrocnemius medialis, whereas the least affected were sartorius, gracilis and the anterior lower leg compartment. Young severely affected patients had a significantly earlier age at disease onset than old mildly affected ones, however no significant differences in disease duration and rate of disease progression were observed. We built two models for FF progression of the thigh and lower leg muscles as two segments adding age at disease onset, baseline MRI data (age, thigh FF% and lower leg FF%) and velocity when performing 6MWT as covariates and obtained two models that predicted FF progression with mean±SD residual values of 2.32×10 −14 ±3.96 and 1.36×10 −15 ±3.96 respectively. In conclusion, this study provides granularity into the way FF progresses over time in the lower limbs. Moreover, the thigh muscles model has a very strong fit with our data and, if validated, could potentially be used in clinical trials to measure effects of upcoming treatments.

Dysferlin plays a key role in cell membrane repair, and its absence or malfunction in patients with dysferlin-deficient limb girdle muscular dystrophy leads to muscle fibre death. Muscle magnetic resonance imaging (MRI) allows non-invasive and repeatable measurements that report on pathological changes observed in dysferlinopathy patients (DP) that lead to muscle dysfunction. We have employed a standardised leg exercise protocol and dynamic 1H- and 23Na-MR imaging and spectroscopy methods to quantify the impact of dysferlin-deficiency on muscle sodium homeostasis. Ten adult DP and 10 age and sex matched healthy volunteers (HV) were recruited and scanned by 23Na and 1H MR imaging and spectroscopy on a 3T clinical MR scanner. Participants performed physiotherapist-guided isometric dorsiflexion contractions to achieve tibialis anterior (TA) muscle exhaustion. Dynamic volume-localised 23Na- and 1H-MR spectroscopy scans were acquired serially for 35 minutes post-exercise. MR data were analysed to determine TA lipid content and the change in TA sodium content, sodium T2* relaxation properties, and TA water 1H T2 relaxation properties. Biexponential analysis of 23Na T2* data was performed to assess dynamic change in the fraction of fast- vs slow-relaxing 23Na. Baseline muscle water 1H T2 and sodium content were significantly higher in DP compared to matched controls. 1H T2 and sodium content dynamics in healthy controls showed post-exercise elevation with slower time-to-peak for sodium content compared to 1H T2. Magnitude of 1H T2 and sodium content post-exercise dynamic change was highly variable in the DP group. Notably, 23Na dynamics in a DP with normal muscle fat fraction were similar to that of the HV group. Biexponential analysis of 23Na T2* showed a post-exercise increase in the slow-relaxing component, consistent with oedematous changes. We have implemented volume-localised 23Na MR spectroscopy for measurement of tissue sodium dynamics with high temporal resolution. Assessment of TA muscle fat fraction, 1H T2, sodium content and sodium T2* relaxation properties revealed differences at baseline and differences in post-exercise dynamics between patients with dysferlinopathy and matched controls. Alterations in sodium content and in post-exercise sodium content and relaxation dynamics in DP may reflect altered ion homeostasis that worsens as chronic muscle damage increases.

The spontaneous contraction of motor units in muscle, i.e. fasciculation, has been recognised as an important diagnostic marker in amyotrophic lateral sclerosis (ALS). Fasciculation can be imaged with a novel MRI technique called motor unit MRI. This technique uses a diffusion weighted sequence on which fasciculation presents as short-living signal voids. We demonstrated an increased fasciculation rate in ALS patients compared to healthy controls by assessing the four body regions relevant in the diagnosis of ALS. The affected body regions differed between patients. This is in line with the heterogeneous disease onset and supports our proposed whole-body approach.

Myostatin is a myokine which acts upon skeletal muscle to inhibit growth and regeneration. Myostatin is endogenously antagonised by follistatin. This study assessed serum myostatin and follistatin concentrations as monitoring or prognostic biomarkers in dysferlinopathy, an autosomal recessively inherited muscular dystrophy. Myostatin was quantified twice with a three-year interval in 76 patients with dysferlinopathy and 38 controls. Follistatin was quantified in 62 of these patients at the same timepoints, and in 31 controls. Correlations with motor function, muscle fat fraction and contractile cross-sectional area were performed. A regression model was used to account for confounding variables. Baseline myostatin, but not follistatin, correlated with baseline function and MRI measures. However, in individual patients, three-year change in myostatin did not correlate with functional or MRI changes. Linear modelling demonstrated that function, serum creatine kinase and C-reactive protein, but not age, were independently related to myostatin concentration. Baseline myostatin concentration predicted loss of ambulation but not rate of change of functional or MRI measures, even when relative inhibition with follistatin was considered. With adjustment for extra-muscular causes of variation, myostatin could form a surrogate measure of functional ability or muscle mass, however myostatin inhibition does not form a promising treatment target in dysferlinopathy.

Sarcopenia is commonly associated with ageing, whereby individuals lose muscle mass and strength. One potential contributor to sarcopenia is the degeneration of motor units (MU), defined as a single motor neuron and the muscle fibres it innervates. We used motor unit MRI (MUMRI) to investigate changes to spontaneous MU activity and morphology in a cohort of healthy ageing adults. We found that MU activity and morphology did not appear to change with healthy ageing. We next aim to apply the MUMRI technique in patients with sarcopenia to look for evidence of accelerated MU loss compared to these healthy controls.

Objectives: Fatigue is one of the most important symptoms needing improvement in Primary Sjogren's syndrome (PSS). Previous data from our group suggest that noninvasive stimulation of the vagus nerve (nVNS) may improve symptoms of fatigue. This experimental medicine study uses the gammaCore device (electroCore) and a sham device to investigate the relationship between nVNS and fatigue in PSS, and to explore potential mechanisms involved. Materials and Methods: Forty participants with PSS were randomly assigned to use active (n = 20) or sham (n = 20) nVNS devices twice daily for 54 days in a double-blind manner. Patient-reported measures of fatigue were collected at baseline and day 56: Profile of Fatigue (PRO-F)-Physical, PRO-F-Mental and Visual Analogue Scale of abnormal fatigue (fVAS). Neurocognitive tests, immunologic responses, electroencephalography alpha reactivity, muscle acidosis, and heart rate variability were compared between devices from baseline to day 56 using analysis of covariance. Results: PRO-F-Physical, PRO-F-Mental, and fVAS scores were significantly reduced at day 56 in the active group only (p = 0.02, 0.02, and 0.04, respectively). Muscle bioenergetics and heart rate variability showed no change between arms. There were significant improvements in digit span and a neurocognitive test (p = 0.03), and upon acute nVNS stimulation, frontal region alpha reactivity showed a significant negative relationship with fatigue scores in the active group (p < 0.01). Conclusions: We observed significant improvements in three measures of fatigue at day 56 with the active device but not the sham device. Directly after device use, fatigue levels correlate with measures of alpha reactivity, suggesting modulation of cholinergic system integrity as a mechanism of action for nVNS.