INTRODUCTION:Accelerated brain aging represents a major feature of multiple sclerosis (MS), and long-term aerobic walking exercise training (ET) may slow brain aging in MS. METHODS:This study examined the effects of 1-year of remotely-delivered aerobic walking ET on brain age in 24 fully-ambulatory persons with MS who underwent 3T MRI. We calculated pre-to-post differences in brain age (derived based on a comparison of participant brain morphology with a large multi-age set of training images using brain-ageR (v2.1)), and compared such changes with the published, annualized rate of accelerated brain aging in relapsing-remitting MS (mean=1.24 years [SD=1.30]), given that 83% of the present sample presented with relapsing-remitting MS. RESULTS:1-year of aerobic walking ET was associated with 0.74 years of brain aging. Such an effect translates into a 0.50-year slowing of brain aging relative to the annualized brain aging rate observed for persons with relapsing-remitting MS (i.e., 1.24 years of brain aging over a 1-year period). CONCLUSIONS:Our preliminary results may support the formal design and test of a larger trial of long-term, remotely-delivered and supported aerobic walking ET on brain age and its downstream consequences as a primary, clinically-relevant outcome in MS.
Sleep disturbance is common in aquaporin-4 immunoglobulin G (AQP4-IgG)-positive neuromyelitis optica spectrum disorder (NMOSD), but its clinical and neurobiological correlates remain incompletely characterized; prior studies used single-modality assessments. In this exploratory cross-sectional study we characterized sleep-wake patterns in clinically stable AQP4-IgG-positive NMOSD and explored volume-sleep associations. Patients and controls underwent questionnaires, 7-day wrist actigraphy and structural MRI; intracranial volume-normalized regional volumes and within-patient volume-sleep associations were examined, with Benjamini-Hochberg false discovery rate (FDR) correction. Twenty-four patients and 26 controls were enrolled; 20 patients and 20 database-derived controls contributed MRI. Patients were older than controls (49.7 versus 33.0 years; g = 1.40); analyses were adjusted for age and body mass index (BMI). After adjustment, patients showed longer time in bed (+58 min) and total sleep time (+50 min), whereas sleep efficiency and wake after sleep onset did not differ. Higher BMI was associated with lower sleep efficiency (r = -0.44) and longer sleep onset latency. Subjective sleep disturbance and pain-related quality of life were worse in NMOSD. After FDR correction, volume reductions were observed in bilateral hippocampal molecular-layer subfields and all five corpus callosum segments, with suggestive thalamic reductions; no volume-sleep association survived. Sleep disturbance in clinically stable AQP4-IgG-positive NMOSD was characterized by greater subjective sleep burden and altered sleep-wake timing, with associations observed with clinical and metabolic factors. These hypothesis-generating findings are consistent with a multifactorial pattern of associations and highlight the need for longitudinal studies integrating objective sleep measurement, neuroimaging and biomarkers.
Background Respiratory dysfunction may occur in multiple sclerosis, and respiratory complications are the most common cause of death. There is limited evidence to guide assessment and management of respiratory dysfunction in people with advanced multiple sclerosis. Objective The aim of this study was to evaluate the assessment and management of respiratory dysfunction in people with advanced multiple sclerosis as part of their routine out-patient care. Methods The medical records of 105 people with multiple sclerosis with Expanded Disability Status Scale (EDSS) score ≥ 8.0 assessed in a multidisciplinary clinic from April 2021 to January 2024 were retrospectively reviewed. Clinical details including age, sex, EDSS score, dysphagia status, respiratory symptoms and planned respiratory interventions were recorded. Results Sixty-four people (61%) reported at least one symptom suggestive of respiratory dysfunction including hypophonia, weak cough, difficulty in breathing, difficulty clearing airway secretions, chest infection and sleep apnoea. Forty-seven people (44.8%) received or were referred for respiratory interventions. Respiratory symptoms and interventions increased with advancing disability, respectively reported in 94% and 76% of those with EDSS ≥9.0. Conclusion Respiratory dysfunction is very common in people with advanced multiple sclerosis and increases with higher EDSS. Screening for respiratory dysfunction can be easily incorporated into routine clinical care by asking about respiratory symptoms. Referral for further assessment and intervention may be required, but this may be constrained by a lack of respiratory services for people with multiple sclerosis.
Background : Artificial intelligence (AI) has been applied across many aspects of multiple sclerosis (MS) theragnostics, from lesion detection in magnetic resonance imaging (MRI), gait assessment and treatment response prediction to drug repurposing. However, a considerable number of models characterized by high technical performance have yet to be translated into a clinical setting. Objective : To investigate the translational barriers between AI-based theragnostic applications and their clinical implementation in MS. Key Messages : Many AI-based models have been trained using retrospective data from a single-center source and are rarely externally validated in larger independent populations. Major barriers to translation include technical issues, limited prospective validation, and multi-modality data, poor interoperability, clinical utility, and ethical concerns. In MS particularly, models must remain reliable across various MS phenotypes and disease modifying therapies making translation even more challenging. Conclusion : High technical performance alone is not sufficient for clinical implementation. Meaningful translation of AI-based theragnostic approaches in MS will require multicenter datasets representative of larger populations, further external validation and transparency. Only through these measures can AI-driven theragnostic approaches evolve from promising research models into clinically meaningful tools for MS care.
BACKGROUND:Multiple sclerosis (MS) is a major cause of chronic neurological disability, and its burden has shifted from premature mortality toward long-term disability. Italy, with its marked north-south gradients and the exceptional epidemiology of Sardinia, offers a unique setting to examine whether this transition has occurred uniformly across regions and sexes. METHODS:Using Global Burden of Disease 2023 subnational estimates, we analysed MS burden in Italy from 1990 to 2023 across five macro-areas. Age-standardized disability-adjusted life years (DALYs), years of life lost (YLLs), years lived with disability (YLDs), and the YLL/YLD ratio were evaluated by sex. Temporal trends were quantified using log-linear regression to estimate annual percentage change (EAPC). Non‑linearities were assessed using Joinpoint regression. We assessed sex-specific differences, changes in proportional composition of DALYs, and heterogeneity of trends across macro-areas. RESULTS:DALY rates increased in all macro-areas, with the steepest rises in Sardinia and South+Sicily, and the smallest increase in the Center. YLDs rose faster than YLLs everywhere, producing a consistent decline in the YLL/YLD ratio and confirming a shift toward disability-dominated burden. EAPCs were positive for DALYs and YLDs in all regions and sexes, highest in Sardinia. Joinpoint regression revealed multiple inflection points. Sex differences were present but smaller than geographic differences. Global interaction tests showed significant heterogeneity across macro-areas for all outcomes. CONCLUSIONS:MS burden in Italy has increased substantially, consistent with a rising population‑level disability burden, which may reflect higher prevalence, longer survival, and evolving epidemiological dynamics. Differences across macro‑areas require continued region‑specific epidemiological monitoring.
