Multiple Sclerosis Misdiagnosis: A Persistent Problem to Solve

AI-based model for automatic identification of multiple sclerosis based on enhanced sea-horse optimizer and MRI scans

This study aims to develop an AI-enhanced methodology for the expedited and accurate diagnosis of Multiple Sclerosis (MS), a chronic disease affecting the central nervous system leading to progressive impairment. Traditional diagnostic methods are slow and require substantial expertise, underscoring the need for innovative solutions. Our approach involves two phases: initially, extracting features from brain MRI images using first-order histograms, the gray level co-occurrence matrix, and local binary patterns. A unique feature selection technique combining the Sine Cosine Algorithm with the Sea-horse Optimizer is then employed to identify the most significant features. Utilizing the eHealth lab dataset, which includes images from 38 MS patients (mean age 34.1 ± 10.5 years; 17 males, 21 females) and matched healthy controls, our model achieved a remarkable 97.97% detection accuracy using the k-nearest neighbors classifier. Further validation on a larger dataset containing 262 MS cases (199 females, 63 males; mean age 31.26 ± 10.34 years) and 163 healthy individuals (109 females, 54 males; mean age 32.35 ± 10.30 years) demonstrated a 92.94% accuracy for FLAIR images and 91.25% for T2-weighted images with the Random Forest classifier, outperforming existing MS detection methods. These results highlight the potential of the proposed technique as a clinical decision-making tool for the early identification and management of MS.

Real-world challenges in the diagnosis of primary progressive multiple sclerosis

BACKGROUND AND PURPOSE

Despite the 2017 revisions to the McDonald criteria, diagnosing primary progressive multiple sclerosis (PPMS) remains challenging. To improve clinical practice, the aim was to identify frequent diagnostic challenges in a real-world setting and associate these with the performance of the 2010 and 2017 PPMS diagnostic McDonald criteria.

METHODS

Clinical, radiological and laboratory characteristics at the time of diagnosis were retrospectively recorded from designated PPMS patient files. Possible complicating factors were recorded such as confounding comorbidity, signs indicative of alternative diagnoses, possible earlier relapses and/or incomplete diagnostic work-up (no cerebrospinal fluid examination and/or magnetic resonance imaging brain and spinal cord). The percentages of patients fulfilling the 2010 and 2017 McDonald criteria were calculated after censoring patients with these complicating factors.

RESULTS

A total of 322 designated PPMS patients were included. Of all participants, it was found that n = 28/322 had confounding comorbidity and/or signs indicative of alternative diagnoses, n = 103/294 had possible initial relapsing and/or uncertainly progressive phenotypes and n = 73/191 received an incomplete diagnostic work-up. When applying the 2010 and 2017 diagnostic PPMS McDonald criteria on n = 118 cases with a full diagnostic work-up and a primary progressive disease course without a better alternative explanation, these were met by 104/118 (88.1%) and 98/118 remaining patients (83.1%), respectively (p = 0.15).

CONCLUSION

Accurate interpretation of the initial clinical course, consideration of alternative diagnoses and a full diagnostic work-up are the cornerstones of a PPMS diagnosis. When these conditions are met, the 2010 and 2017 McDonald criteria for PPMS perform similarly, emphasizing the importance of their appropriate application in clinical practice.

Comprehensible Machine-Learning-Based Models for the Pre-Emptive Diagnosis of Multiple Sclerosis Using Clinical Data: A Retrospective Study in the Eastern Province of Saudi Arabia

Multiple Sclerosis (MS) is characterized by chronic deterioration of the nervous system, mainly the brain and the spinal cord. An individual with MS develops the condition when the immune system begins attacking nerve fibers and the myelin sheathing that covers them, affecting the communication between the brain and the rest of the body and eventually causing permanent damage to the nerve. Patients with MS (pwMS) might experience different symptoms depending on which nerve was damaged and how much damage it has sustained. Currently, there is no cure for MS; however, there are clinical guidelines that help control the disease and its accompanying symptoms. Additionally, no specific laboratory biomarker can precisely identify the presence of MS, leaving specialists with a differential diagnosis that relies on ruling out other possible diseases with similar symptoms. Since the emergence of Machine Learning (ML) in the healthcare industry, it has become an effective tool for uncovering hidden patterns that aid in diagnosing several ailments. Several studies have been conducted to diagnose MS using ML and Deep Learning (DL) models trained using MRI images, achieving promising results. However, complex and expensive diagnostic tools are needed to collect and examine imaging data. Thus, the intention of this study is to implement a cost-effective, clinical data-driven model that is capable of diagnosing pwMS. The dataset was obtained from King Fahad Specialty Hospital (KFSH) in Dammam, Saudi Arabia. Several ML algorithms were compared, namely Support Vector Machine (SVM), Decision Tree (DT), Logistic Regression (LR), Random Forest (RF), Extreme Gradient Boosting (XGBoost), Adaptive Boosting (AdaBoost), and Extra Trees (ET). The results indicated that the ET model outpaced the rest with an accuracy of 94.74%, recall of 97.26%, and precision of 94.67%.

Multiple sclerosis diagnostic delay and its associated factors in Upper Egyptian patients

The earlier the diagnosis of multiple sclerosis (MS), the sooner disease-modifying treatments can be initiated. However, significant delays still occur in developing countries. We aimed to identify factors leading to delayed diagnosis of MS in Upper Egypt. One hundred forty-two patients with remitting relapsing MS (RRMS) were recruited from 3 MS units in Upper Egypt. Detailed demographic and clinical data were collected. Neurological examination and assessment of the Disability Status Scale (EDSS) were performed. The mean age was 33.52 ± 8.96 years with 72.5% of patients were females. The mean time from symptom onset to diagnosis was 18.63 ± 27.87 months and the median was 3 months. Seventy-two patients (50.7%) achieved diagnosis within three months after the first presenting symptom (early diagnosis), while seventy patients (49.3%) had more than three months delay in diagnosis (delayed diagnosis). Patients with a delayed diagnosis frequently presented in the period before 2019 and had a significantly higher rate of initial non-motor presentation, initial non-neurological consultations, prior misdiagnoses, and a higher relapse rate. Another possible factor was delayed MRI acquisition following the initial presentation in sixty-six (46.5%) patients. Multivariable logistic regression analysis demonstrated that earlier presentation, initial non-neurological consultation, and prior misdiagnosis were independent predictors of diagnostic delay. Despite advances in MS management in Egypt, initial non-neurological consultation and previous misdiagnoses are significant factors responsible for delayed diagnosis in Upper Egypt.

The borderland of multiple sclerosis and functional neurological disorder: A call for clinical research and vigilance

BACKGROUND AND PURPOSE

Functional neurological disorders (FNDs) have attracted much attention from the neurological medical community over the last decades as new developments in neurosciences have reduced stigma around these by showing brain network dysfunctions. An overlap with other neurological conditions such as multiple sclerosis (MS) is well known by clinicians but there is a lack of clinical and fundamental research in this field to better define diagnosis and therapeutic decisions, as well as a lack of deep understanding of the underlying pathophysiology.

AIM

We aimed to provide a critical commentary on the state of knowledge about the borderland between FNDs and MS.

METHODS

We based our commentary on a joint point of view between an FND specialist and an MS expert.

RESULTS

A brief review of the previous literature and relevant new studies covering the overlap between FNDs and MS is presented, along with suggestions for future research directions.

CONCLUSION

There are clear diagnostic criteria for both FNDs and MS and a strict application of these will help better diagnosis and prevent unnecessary treatment escalation in MS or absence of referral to multimodal therapy in FND. Better teaching of younger neurologists is needed as well as prospective research focusing on pathophysiology.

An area of greatest vulnerability - recent advances in kidney injury

In this issue of the Biomedical Journal the reader is provided with an insight into the latest observations and advances in acute kidney injury as well as chronic kidney disease. The current SARS-CoV-2 variants are reviewed, and the role of long non-coding RNA in HIV therapy is explored. Furthermore, the potential of metabolomics as means to diagnose multiple sclerosis as well as tuberculosis is presented. Other topics of this issue include the restoration of the spermatogonial stem cell niche; atherosclerosis and the use of improved ultrasound images; and the effect of transcranial magnetic stimulation in patients with autism spectrum disorder. Finally, it is shown how continuous passive motion can be used as supportive therapeutic approach in children with cerebral palsy, and minimally invasive surgery is presented as valid alternative in cases of spine metastasis.

Immunopathology of the Optic Nerve in Multiple Sclerosis

Optic neuritis, a primary clinical manifestation commonly observed in multiple sclerosis (MS) is a major factor leading to permanent loss of vision. Despite decreased vision (optic neuritis), diplopia, and nystagmus, the immunopathology of the optic nerve in MS is unclear. Here, we have characterised the optic nerve pathology in a large cohort of MS cases (n=154), focusing on the immune responses in a sub-cohort of MS (n=30) and control (n=6) cases. Immunohistochemistry was used to characterise the myeloid (HLA-DR, CD68, Iba1, TMEM119, P2RY12) and adaptive immune cells (CD4, CD8, CD138) in the parenchyma, perivascular spaces, and meninges in optic nerve tissues from MS and control cases. Of the 154 MS cases, 122 (79%) reported visual problems of which 99 (81%) optic nerves showed evidence of damage. Of the 31 cases with no visual disturbances, 19 (61%) showed evidence of pathology. A pattern of myeloid cell activity and demyelination in the optic nerve was similar to white matter lesions in the brain and spinal cord. In the optic nerves, adaptive immune cells were more abundant in the meninges close to active and chronic active lesions, and significantly higher compared to the parenchyma. Similar to brain tissues in this Dutch cohort, B-cell follicles in the meninges were absent. Our study reveals that optic nerve pathology is a frequent event in MS and may occur in the absence of clinical symptoms.

Determination of CSF GFAP, CCN5, and vWF Levels Enhances the Diagnostic Accuracy of Clinically Defined MS From Non-MS Patients With CSF Oligoclonal Bands

Background

Inclusion of cerebrospinal fluid (CSF) oligoclonal IgG bands (OCGB) in the revised McDonald criteria increases the sensitivity of diagnosis when dissemination in time (DIT) cannot be proven. While OCGB negative patients are unlikely to develop clinically definite (CD) MS, OCGB positivity may lead to an erroneous diagnosis in conditions that present similarly, such as neuromyelitis optica spectrum disorders (NMOSD) or neurosarcoidosis.

Objective

To identify specific, OCGB-complementary, biomarkers to improve diagnostic accuracy in OCGB positive patients.

Methods

We analysed the CSF metabolome and proteome of CDMS (n=41) and confirmed non-MS patients (n=64) comprising a range of CNS conditions routinely encountered in neurology clinics.

Results

OCGB discriminated between CDMS and non-MS with high sensitivity (85%), but low specificity (67%), as previously described. Machine learning methods revealed CCN5 levels provide greater accuracy, sensitivity, and specificity than OCGB (79%, +5%; 90%, +5%; and 72%, +5% respectively) while glial fibrillary acidic protein (GFAP) identified CDMS with 100% specificity (+33%). A multiomics approach improved accuracy further to 90% (+16%).

Conclusion

The measurement of a few additional CSF biomarkers could be used to complement OCGB and improve the specificity of MS diagnosis when clinical and radiological evidence of DIT is absent.

The frequency and characteristics of multiple sclerosis misdiagnosis in Latin America: A referral center study in Buenos Aires, Argentina

OBJECTIVE

Most contemporary data concerning the frequency and causes of multiple sclerosis (MS) misdiagnosis are from North America and Europe with different healthcare system structure and resources than countries in Latin America. We sought to determine the frequency, and potential contributors to MS misdiagnosis in patients evaluated at an MS referral center in Argentina.

METHODS

The study was a retrospective medical record review. We included patients evaluated at the MS Clinic at Fleni between April 2013 and March 2021. Diagnoses prior to consultation, final diagnoses after consultation, demographic, clinical and paraclinical data, and treatment were extracted and classified.

RESULTS

Seven hundred thirty-six patients were identified. Five hundred seventy-two presented with an established diagnosis of MS and after evaluation, misdiagnosis was identified in 89 (16%). Women were at 83% greater risk of misdiagnosis (p = 0.034). The most frequent alternative diagnoses were cerebrovascular disease, radiological isolated syndrome (RIS), and headache. Seventy-four (83%) of misdiagnosed patients presented with a syndrome atypical for demyelination, 62 (70%) had an atypical brain magnetic resonance imaging (MRI), and 54 (61%) were prescribed disease-modifying therapy.

CONCLUSION

Sixteen percent of patients with established MS were subsequently found to have been misdiagnosed. Women were at higher risk for misdiagnosis. Expert application of the McDonald criteria may prevent misdiagnosis and its associated morbidity and healthcare system cost.

Th17-Related Cytokines as Potential Discriminatory Markers between Neuromyelitis Optica (Devic's Disease) and Multiple Sclerosis-A Review

Multiple sclerosis (MS) and Devic’s disease (NMO; neuromyelitis optica) are autoimmune, inflammatory diseases of the central nervous system (CNS), the etiology of which remains unclear. It is a serious limitation in the treatment of these diseases. The resemblance of the clinical pictures of these two conditions generates a partial possibility of introducing similar treatment, but on the other hand, a high risk of misdiagnosis. Therefore, a better understanding and comparative characterization of the immunopathogenic mechanisms of each of these diseases are essential to improve their discriminatory diagnosis and more effective treatment. In this review, special attention is given to Th17 cells and Th17-related cytokines in the context of their potential usefulness as discriminatory markers for MS and NMO. The discussed results emphasize the role of Th17 immune response in both MS and NMO pathogenesis, which, however, cannot be considered without taking into account the broader perspective of immune response mechanisms.

Pathology-supported genetic testing as a method for disability prevention in multiple sclerosis (MS). Part I. Targeting a metabolic model rather than autoimmunity

In this Review (Part I), we investigate the scientific evidence that multiple sclerosis (MS) is caused by the death of oligodendrocytes, the cells that synthesize myelin, due to a lack of biochemical and nutritional factors involved in mitochondrial energy production in these cells. In MS, damage to the myelin sheaths surrounding nerve axons causes disruption of signal transmission from the brain to peripheral organs, which may lead to disability. However, the extent of disability is not deterred by the use of MS medication, which is based on the autoimmune hypothesis of MS. Rather, disability is associated with the loss of brain volume, which is related to the loss of grey and white matter. A pathology-supported genetic testing (PSGT) method, developed for personalized assessment and treatment to prevent brain volume loss and disability progression in MS is discussed. This involves identification of MS-related pathogenic pathways underpinned by genetic variation and lifestyle risk factors that may converge into biochemical abnormalities associated with adverse expanded disability status scale (EDSS) outcomes and magnetic resonance imaging (MRI) findings during patient follow-up. A Metabolic Model is presented which hypothesizes that disability may be prevented or reversed when oligodendrocytes are protected by nutritional reserve. Evidence for the validity of the Metabolic Model may be evaluated in consecutive test cases following the PSGT method. In Part II of this Review, two cases are presented that describe the PSGT procedures and the clinical outcomes of these individuals diagnosed with MS.

Beyond McDonald: updated perspectives on MRI diagnosis of multiple sclerosis

INTRODUCTION

Magnetic resonance imaging (MRI) is an essential paraclinical test to establish an accurate and early diagnosis of multiple sclerosis (MS), which is based on the application of the McDonald criteria.

AREAS COVERED

The objective of this article is to analyze, based on publicly available database since the publication of the 2017 McDonald diagnostic criteria, the clinical impact of these criteria, to discuss the potential inclusion within these criteria of the optic nerve to demonstrate dissemination in space, and to guide the acquisition and interpretation of MRI scans for diagnostic purposes. Finally, the authors will review emerging MRI features that could improve the specificity of MRI in the diagnosis of MS and consequently minimize the misdiagnosis of this disease.

EXPERT OPINION

Although the optic nerve has not been included as one of the topographies required to demonstrate demyelinating lesion disseminated in space in the 2017 McDonald criteria, new studies seem to show some improvement in the sensitivity of these criteria when this topography is considered. New radiological findings such as the central vein sign and iron rims, should be considered within the typical MRI features of this disease with the objective of minimizing MRI-based diagnostic errors.

Proteomics of Multiple Sclerosis: Inherent Issues in Defining the Pathoetiology and Identifying (Early) Biomarkers

Multiple Sclerosis (MS) is a demyelinating disease of the human central nervous system having an unconfirmed pathoetiology. Although animal models are used to mimic the pathology and clinical symptoms, no single model successfully replicates the full complexity of MS from its initial clinical identification through disease progression. Most importantly, a lack of preclinical biomarkers is hampering the earliest possible diagnosis and treatment. Notably, the development of rationally targeted therapeutics enabling pre-emptive treatment to halt the disease is also delayed without such biomarkers. Using literature mining and bioinformatic analyses, this review assessed the available proteomic studies of MS patients and animal models to discern (1) whether the models effectively mimic MS; and (2) whether reasonable biomarker candidates have been identified. The implication and necessity of assessing proteoforms and the critical importance of this to identifying rational biomarkers are discussed. Moreover, the challenges of using different proteomic analytical approaches and biological samples are also addressed.

The frequency and characteristics of MS misdiagnosis in patients referred to the multiple sclerosis centre of Catalonia

Background:

Multiple sclerosis (MS) misdiagnosis may cause physical and emotional damage to patients.

Objectives:

The objective of this study is to determine the frequency and characteristics of MS misdiagnosis in patients referred to the Multiple Sclerosis Centre of Catalonia.

Methods:

We designed a prospective study including all new consecutive patients referred to our centre between July 2017 and June 2018. Instances of misdiagnosis were identified, and referral diagnosis and final diagnosis were compared after 1 year of follow-up. Association of misdiagnosis with magnetic resonance imaging (MRI) findings, presence of comorbidities and family history of autoimmunity were assessed.

Results:

A total of 354 patients were referred to our centre within the study period, 112 (31.8%) with ‘established MS’. Misdiagnosis was identified in eight out of 112 cases (7.1%). MRI identified multifocal white matter lesions, deemed non-specific or not suggestive of MS in all misdiagnosed cases. Patients with MS misdiagnosis had more comorbidities in general than patients with MS ( p = 0.026) as well as a personal history of autoimmunity ( p < 0.001).

Conclusion:

A low frequency of MS misdiagnosis was found in our clinical setting. Multifocal non-specific white matter lesions in referral MRI examinations and the presence of comorbidities, including a personal history of autoimmunity, seem to be contributing factors to misdiagnosis.

Biomechanical muscle stiffness measures of extensor digitorum explain potential mechanism of McArdle sign

BACKGROUND

McArdle sign is a phenomenon of impaired gait and muscle weakness that occurs with neck flexion, immediately reversible with neck extension. A recent report measured the specificity of this sign for multiple sclerosis by measuring differences in peak torque of the extensor digitorum between neck extension and flexion.

METHODS

This substudy included 73 participants (29 multiple sclerosis, 20 non-multiple sclerosis myelopathies, 5 peripheral nerve disorders, and 19 healthy controls). The effect of neck position was assessed on muscle stiffness and neuromechanical error of the extensor digitorum.

FINDINGS

Patients with multiple sclerosis had greater neuromechanical error (sum of squared error of prediction) compared to controls (P = 0.023) and non-multiple sclerosis myelopathies (P = 0.003). Neuromechanical error also provided improved sensitivity/specificity of McArdle sign. Peak torque, muscle stiffness, and neuromechanical error could distinguish multiple sclerosis from other myelopathies with 80% specificity and 97% sensitivity (AUC = 0.95).

INTERPRETATION

A decrease in muscle stiffness and neuromechanical error in neck flexion compared to extension are additional indicators for a diagnosis of multiple sclerosis. Analysis of muscle stiffness may provide insights into the pathophysiology of this specific clinical sign for multiple sclerosis. Furthermore, muscle stiffness may provide an additional accurate, simple assessment to evaluate multiple sclerosis therapeutic interventions and disease progression.

An emerging potential of metabolomics in multiple sclerosis: a comprehensive overview

Multiple sclerosis (MS) is an inflammatory demyelinating disease of the nervous system that primarily affects young adults. Although the exact etiology of the disease remains obscure, it is clear that alterations in the metabolome contribute to this process. As such, defining a reliable and disease-specific metabolome has tremendous potential as a diagnostic and therapeutic strategy for MS. Here, we provide an overview of studies aimed at identifying the role of metabolomics in MS. These offer new insights into disease pathophysiology and the contributions of metabolic pathways to this process, identify unique markers indicative of treatment responses, and demonstrate the therapeutic effects of drug-like metabolites in cellular and animal models of MS. By and large, the commonly perturbed pathways in MS and its preclinical model include lipid metabolism involving alpha-linoleic acid pathway, nucleotide metabolism, amino acid metabolism, tricarboxylic acid cycle, d-ornithine and d-arginine pathways with collective role in signaling and energy supply. The metabolomics studies suggest that metabolic profiling of MS patient samples may uncover biomarkers that will advance our understanding of disease pathogenesis and progression, reduce delays and mistakes in diagnosis, monitor the course of disease, and detect better drug targets, all of which will improve early therapeutic interventions and improve evaluation of response to these treatments.