Overview

The Revised McDonald Criteria (2017), published by Thompson, Banwell, Barkhof, and colleagues in Lancet Neurology in January 2018, represent the internationally endorsed standard for the diagnosis of multiple sclerosis (MS). The criteria formalise the fundamental diagnostic requirement that MS lesions must be disseminated across the central nervous system in both space (DIS) and time (DIT), and that no alternative diagnosis better explains the clinical and radiological findings.

The 2017 revision updated the previous 2010 McDonald Criteria (Polman et al.) in three clinically significant ways: the acceptance of CSF-specific oligoclonal bands (OCBs) as a substitute for DIT; the inclusion of symptomatic lesions as valid contributors to DIS and DIT; and the addition of cortical lesions alongside juxtacortical lesions as one of five recognised CNS regions for DIS. These changes were designed to allow earlier diagnosis without sacrificing diagnostic specificity, enabling timely access to disease-modifying therapy while reducing the window of diagnostic uncertainty that carries its own psychological and therapeutic costs.

The criteria apply primarily to patients presenting with a typical clinically isolated syndrome (CIS) or insidious neurological progression suggestive of primary progressive MS (PPMS). They are not designed for atypical presentations in which alternative inflammatory CNS disorders, including neuromyelitis optica spectrum disorder (NMOSD) and MOG antibody-associated disease (MOGAD), must be actively excluded before McDonald criteria are applied.

Multiple Sclerosis: Disease Biology and Epidemiology

Multiple sclerosis is a chronic immune-mediated inflammatory demyelinating and neurodegenerative disease of the central nervous system. It is the leading cause of non-traumatic neurological disability in young adults in high-income countries, with a global prevalence of approximately 2.8 million people. The disease is approximately two to three times more prevalent in women than in men and has a peak age of onset between 20 and 40 years, though paediatric and late-onset presentations occur.

The pathological hallmark of MS is the demyelinating plaque: a focal area of myelin loss, axonal damage, astrocytic gliosis, and variable remyelination within the white and grey matter of the brain, optic nerves, and spinal cord. Active plaques show perivenular lymphocytic and macrophage infiltration driven by autoreactive T-cells directed against myelin antigens, breach of the blood-brain barrier (manifest as gadolinium enhancement on MRI), and ongoing oligodendrocyte injury. Over time, progressive neuroaxonal loss accumulates in the absence of overt new inflammatory activity, producing the neurodegeneration that underlies progressive disability accrual independent of relapse frequency.

The majority of patients (approximately 85%) present with relapsing-remitting MS (RRMS), characterised by discrete episodes of neurological dysfunction (attacks or relapses) followed by complete or partial recovery. Most untreated RRMS patients eventually transition to secondary progressive MS (SPMS), characterised by steady neurological worsening independent of relapses. Approximately 10 to 15% of patients have primary progressive MS (PPMS) from onset, with insidious neurological deterioration without clear relapses, most commonly manifesting as a progressive myelopathy.

The geographic distribution of MS prevalence increases with distance from the equator, with high-prevalence regions including Northern Europe, North America, Australia, and New Zealand. This gradient is partially explained by vitamin D insufficiency related to reduced sun exposure, and partially by genetic susceptibility concentrated in populations of Northern European ancestry, particularly HLA-DRB1*15:01 haplotype carriage.

The Core Diagnostic Principles: DIS and DIT

The McDonald Criteria are built on two foundational concepts that define the unique spatiotemporal pattern of MS-related inflammatory demyelination:

Dissemination in Space (DIS)

DIS requires evidence that demyelinating lesions are present in multiple anatomically distinct regions of the CNS, reflecting the characteristic multifocal nature of MS pathology. In clinical terms, DIS can be demonstrated by:

• Two or more clinical attacks affecting distinct neurological functions (implying lesions in at least two separate anatomical locations), with or without corroborating MRI.
• MRI evidence of one or more T2-hyperintense lesions characteristic of MS in at least two of the five recognised CNS regions (see below).

The 2017 criteria define five CNS regions for MRI-based DIS. A single T2 lesion in each qualifying region is sufficient; there is no minimum lesion count per region beyond one:

1. Periventricular: Lesions directly abutting the lateral ventricles, including the corpus callosum. The classic Dawson’s fingers appearance on sagittal MRI (ovoid periventricular lesions oriented perpendicular to the long axis of the corpus callosum) is characteristic of MS.
2. Cortical or juxtacortical: Lesions within the cerebral cortex or in the white matter immediately adjacent to it (juxtacortical). Cortical lesions were added to the 2017 criteria to reflect the sensitivity improvements of modern MRI sequences (FLAIR, DIR, phase-sensitive inversion recovery) for detecting intracortical and leukocortical lesions. Cortical lesions are increasingly recognised as clinically significant contributors to cognitive impairment and disability in MS.
3. Infratentorial: Lesions in the brainstem (midbrain, pons, medulla oblongata) or cerebellum. Brainstem and cerebellar lesions are typical MS locations and produce a characteristic pattern of diplopia, internuclear ophthalmoplegia, vertigo, and ataxia when symptomatic.
4. Spinal cord: Intramedullary T2-hyperintense lesions within the spinal cord. MS spinal cord lesions are typically short-segment (less than two vertebral levels in length), peripheral, and located in the posterior or lateral columns, distinguishing them from the long, central, T2-bright lesions of neuromyelitis optica spectrum disorder.
5. Optic nerve: Included by the MAGNIMS (Magnetic Resonance Imaging in MS) study group as a fifth region and applied in clinical practice at many centres, though not formally incorporated into the 2017 core McDonald Criteria text. Optic neuritis, one of the most common CIS presentations, produces T2 signal abnormality and gadolinium enhancement of the optic nerve on dedicated fat-saturated sequences.

A critical 2017 update is that both symptomatic and asymptomatic lesions count toward DIS (and DIT). In the 2010 criteria, symptomatic lesions in certain presentations could not contribute to DIS. This restriction was removed in 2017 based on evidence that symptomatic lesions are pathologically identical to asymptomatic lesions and carry equivalent diagnostic information.

Dissemination in Time (DIT)

DIT requires evidence that demyelinating lesions have occurred at different points in time, reflecting the relapsing or progressive temporal pattern of MS and distinguishing it from monophasic inflammatory conditions such as acute disseminated encephalomyelitis (ADEM). DIT can be demonstrated by any one of four methods:

1. Simultaneous gadolinium-enhancing and non-enhancing T2 lesions on a single MRI: The co-existence of enhancing lesions (reflecting acute blood-brain barrier disruption, typically less than 6 weeks old) and non-enhancing T2 lesions (reflecting prior inflammatory events) on one scan provides intrinsic temporal evidence of lesions at different ages, enabling single-scan DIT demonstration.
2. A new T2 or gadolinium-enhancing lesion on a follow-up MRI compared to a baseline scan, regardless of the time interval between scans. Even a follow-up MRI obtained within days to weeks of baseline can fulfill DIT if a new lesion is identified.
3. A second clinical attack separated from the first by at least 30 days and involving a different neurological symptom or sign (implying a lesion in a distinct CNS location). Two qualifying clinical attacks are the most robust demonstration of DIT and require no MRI or CSF corroboration in patients who also have two or more objective clinical lesions.
4. CSF-specific oligoclonal bands (OCBs) detected on cerebrospinal fluid analysis. This is the principal new addition to the 2017 criteria and is discussed in detail below.

CSF Oligoclonal Bands as a DIT Substitute

The 2017 McDonald Criteria introduced CSF-specific oligoclonal IgG bands as an acceptable substitute for DIT, the most impactful single change from the 2010 version. This change was based on substantial evidence that OCBs are a robust marker of intrathecal chronic immune activation in MS, with high sensitivity for established MS and strong predictive value for conversion from CIS to clinically definite MS.

Biology of CSF Oligoclonal Bands

Oligoclonal bands represent discrete immunoglobulin G fractions produced by clonally restricted B-cell populations within the CNS compartment. Their detection in CSF but not in a simultaneously obtained serum sample (CSF-restricted OCBs) indicates intrathecal IgG synthesis, i.e. local antibody production by plasmablasts and plasma cells that have migrated across the blood-brain barrier and become resident within the CNS, meninges, or perivascular spaces.

The antigens targeted by OCB-producing B-cells in MS remain incompletely characterised, though some studies have identified antibodies against Epstein-Barr nuclear antigen (EBNA-1), spermidine synthase, and other targets. The persistence of OCBs throughout the disease course in most patients, largely independent of clinical activity or treatment, suggests that they reflect a stable intrathecal immune network rather than a dynamic inflammatory marker.

Detection and Interpretation

The gold standard for OCB detection is isoelectric focusing (IEF) with immunodetection performed on paired CSF and serum samples. This technique separates IgG molecules by charge and detects discrete band patterns visible on gel or membrane. A positive result for the McDonald Criteria requires:

• Two or more IgG bands present in the CSF that are absent in the paired serum sample (type 2 or type 3 pattern by Reiber-Felgenhauer classification).
• The bands must be CSF-specific, not merely present in both CSF and serum (which would indicate systemic immunoglobulin production, type 4 or 5, and does not satisfy the criterion).

OCBs are present in approximately 85 to 95% of patients with clinically definite MS, making them the most sensitive non-MRI biomarker in the disease. CIS patients with positive OCBs are more than twice as likely to develop a second clinical attack and convert to CDMS compared with OCB-negative CIS patients, a finding consistent across multiple prospective cohort studies. The negative predictive value of OCBs (absence of OCBs making MS less likely) is less robust; OCB-negative MS exists and is particularly associated with late-onset MS and monosymptomatic optic neuritis in some series.

The use of OCBs as a DIT substitute applies specifically to patients where DIS has already been demonstrated clinically or by MRI. OCBs alone, without DIS, do not constitute an MS diagnosis, as intrathecal IgG synthesis is not specific to MS and occurs in other CNS infections, inflammatory, and autoimmune conditions.

Additional CSF Findings

While only OCBs are incorporated into the 2017 McDonald Criteria as a diagnostic variable, several other CSF findings support an MS diagnosis and help exclude alternative diagnoses:

• IgG index: The ratio of CSF IgG to albumin divided by the corresponding serum ratio, quantifying the degree of intrathecal IgG synthesis. An elevated IgG index (>0.7 or >0.77 depending on laboratory method) corroborates intrathecal immune activation but is less sensitive than OCBs for MS.
• Kappa free light chains (kFLC): An emerging biomarker with sensitivity and specificity for MS comparable to OCBs in some studies, measurable by automated nephelometry without the technical expertise required for IEF. Not yet formally incorporated into McDonald Criteria but referenced in ECTRIMS and MAGNIMS guidance documents.
• Cell count and protein: Most MS patients have a normal or mildly elevated CSF white cell count (typically less than 50 cells/μL, predominantly lymphocytes) and normal or mildly elevated total protein. A CSF pleocytosis greater than 50 cells/μL should prompt investigation for alternative diagnoses including CNS infection, CNS lymphoma, or sarcoidosis.
• Neurofilament light chain (NfL): A marker of axonal injury released during neuroaxonal damage, detectable in both CSF and serum. Elevated in active MS and useful for monitoring treatment response and disease activity, but not currently part of diagnostic criteria.

Diagnostic Criteria by Clinical Presentation

Presentation 1: Two or More Attacks, Two or More Objective Lesions

This presentation requires no additional data for an MS diagnosis. Two or more discrete clinical attacks at different points in time (fulfilling DIT clinically) involving objective neurological findings in at least two anatomically distinct locations (fulfilling DIS clinically) constitute sufficient evidence under the McDonald Criteria. No MRI, lumbar puncture, or additional testing is required to make the diagnosis, though neuroimaging is standard practice for baseline assessment, exclusion of mimics, and treatment decision-making.

The requirement for objective clinical evidence of lesions means that the neurological examination must demonstrate findings consistent with the symptomatic complaint (e.g. an afferent pupillary defect corroborating optic neuritis, a sensory level corroborating myelitis). Purely subjective symptoms without objective examination findings do not constitute objective evidence of a lesion. However, historical symptoms can provide additional contextual evidence if typical of MS and the clinician judges them plausible even without contemporaneous examination records.

Presentation 2: Two or More Attacks, One Objective Lesion

In this scenario, DIT is already satisfied clinically (two distinct temporal episodes), but DIS requires additional demonstration because only one objective lesion is documented on examination. DIS can be met by:

• MRI demonstrating T2 lesions in two or more of the five characteristic CNS areas, or
• A future clinical attack implicating a different CNS site (providing both DIS and confirming DIT clinically).

This presentation commonly arises when a second relapse occurs at the same anatomical location as the first (e.g. two episodes of optic neuritis in the same eye), yielding two temporal episodes but only one objectively examined lesion location. MRI of the brain and spinal cord is the practical path to demonstrating DIS without awaiting a third clinical episode.

Critically, the 2017 criteria allow historical evidence of a prior attack in a different location to be counted, even if not contemporaneously examined, provided the clinician regards it as plausible on the basis of symptom description, onset, duration, and recovery pattern. This flexibility accelerates diagnosis in patients who present to neurology after their second event but recall a prior episode that was never formally evaluated.

Presentation 3: One Attack (CIS), Two or More Objective Lesions

A single clinical attack with two or more objective lesions satisfies DIS clinically. The remaining requirement is DIT. Any one of the four DIT methods is acceptable:

• Simultaneous enhancing and non-enhancing lesions on MRI at the time of presentation
• A new T2 or enhancing lesion on follow-up MRI
• A second clinical attack (at least 30 days after the first)
• CSF-specific oligoclonal bands

This is one of the most common scenarios in clinical practice where CSF oligoclonal bands provide diagnostic value. A patient presenting with a first clinical episode (for example, partial transverse myelitis) who has two spinal cord lesions and two brain lesions on MRI (satisfying DIS clinically or by MRI) but no gadolinium enhancement on the acute scan and no prior MRI for comparison can receive an MS diagnosis immediately if lumbar puncture demonstrates CSF-specific OCBs, without waiting for a second attack or repeat MRI.

Presentation 4: One Attack (CIS), One Objective Lesion

The most common and diagnostically challenging scenario is a patient presenting with their first neurological event and only one lesion on examination. This is the classic clinically isolated syndrome (CIS) presentation. Both DIS and DIT must be demonstrated by non-clinical means:

• DIS: MRI showing T2 lesions in two or more of the five CNS regions, or a future clinical attack at a different location.
• DIT: Any of the four DIT methods (simultaneous lesions, new lesion on follow-up MRI, second clinical attack, or CSF OCBs).

In practical terms, a patient presenting with a typical CIS (e.g. unilateral optic neuritis) who has characteristic periventricular, juxtacortical, and infratentorial T2 lesions on brain MRI (satisfying DIS by MRI) and positive CSF OCBs (satisfying DIT by OCBs) can receive an MS diagnosis at the time of their first clinical event, without waiting for a second attack or follow-up MRI demonstrating disease activity. This represents the most clinically impactful application of the 2017 criteria, as it may compress the time to diagnosis and treatment initiation by months to years compared with the 2010 criteria in OCB-positive patients.

Presentation 5: Insidious Progression (Primary Progressive MS)

Primary progressive MS requires a fundamentally different diagnostic approach because discrete clinical attacks are absent by definition. The 2017 McDonald PPMS criteria require:

• One year of disability progression (retrospective or prospective), independent of relapses. Progression must be continuous, not relapse-driven, and affect walking ability, upper limb function, or visual and cognitive function in ways consistent with a spinal cord or brain origin.
• Plus at least 2 of the following 3 criteria:
  1. Brain DIS: One or more T2 lesions in at least one of the three characteristic brain regions (periventricular, cortical/juxtacortical, or infratentorial). Note: for PPMS, the brain DIS criterion requires only one qualifying brain region, not two as required for relapsing-onset DIS.
  2. Spinal cord DIS: Two or more T2 spinal cord lesions with characteristics typical of MS (short-segment, peripheral, posterior or lateral column predominant).
  3. CSF-specific oligoclonal bands.

PPMS most commonly presents as a progressive spastic paraparesis in patients aged 40 to 60 years. The differential diagnosis is broad and includes hereditary spastic paraparesis, dural arteriovenous fistula, primary lateral sclerosis, B12 deficiency myelopathy, copper deficiency myelopathy, cervical spondylotic myelopathy, and HTLV-1 associated myelopathy. Comprehensive neuroimaging (brain and full spinal cord MRI), CSF analysis, and metabolic and infective screening are mandatory before applying PPMS criteria. The requirement for two of three supportive criteria (compared to just one lesion region in the 2010 criteria for PPMS) was maintained in 2017 to preserve specificity in this challenging differential diagnosis context.

Historical Evolution of the McDonald Criteria

2001 Criteria (McDonald et al.)

The original McDonald Criteria, published in Annals of Neurology in 2001, represented a landmark advance by formally incorporating MRI findings alongside clinical evidence for the first time, replacing the purely clinical Poser Criteria (1983) that required two separate clinical episodes with objective signs for a definite MS diagnosis. The 2001 criteria introduced specific MRI requirements for DIS and DIT, including requirements for the number and location of T2 lesions and specified intervals for follow-up MRI to demonstrate a new T2 lesion (at least 30 days after the index event).

2005 Revision (Polman et al.)

The 2005 revision simplified the MRI requirements for DIT. It clarified that a single new T2 lesion on follow-up MRI at any time after a baseline scan (removing the 30-day minimum interval for the follow-up scan) was sufficient for DIT. The DIS MRI criteria were also refined, reducing the minimum number of T2 lesions required from the Barkhof/Tintore criteria of three or four lesions to a simpler formulation.

2010 Revision (Polman et al.)

The 2010 criteria introduced the most significant simplification to that point: the concept of demonstrating DIT on a single MRI scan by the simultaneous presence of gadolinium-enhancing and non-enhancing T2 lesions. This eliminated the need for follow-up MRI in patients whose initial scan showed both active and inactive lesions. The DIS requirement was also simplified to one or more T2 lesions in at least two of four CNS locations (periventricular, juxtacortical, infratentorial, spinal cord), reducing the previous multi-lesion threshold. These changes substantially increased the proportion of CIS patients who could receive an immediate MS diagnosis.

2017 Revision (Thompson et al.)

The 2017 revision preserved the 2010 framework while adding three key changes:

1. CSF-specific OCBs as a substitute for DIT, enabling diagnosis at CIS presentation in OCB-positive patients with established DIS by MRI.
2. Symptomatic lesions permitted to count toward DIS and DIT, removing the 2010 restriction.
3. Cortical lesions added to juxtacortical in the DIS criteria, expanding the recognition of intracortical and leukocortical lesions detectable on advanced MRI sequences. The five-region DIS criterion (versus the four-region 2010 model) reflects the addition of this cortical/juxtacortical category as a distinct entity.

Validation studies comparing the 2017 to the 2010 criteria report that approximately 13 to 20% more CIS patients receive an MS diagnosis at first presentation under the 2017 criteria, with maintained specificity of approximately 95 to 98% when strictly applied to typical CIS with alternative diagnoses excluded.

MRI Protocol Considerations

The accuracy and reproducibility of McDonald Criteria application depends critically on the quality of the MRI acquisition. Brain and spinal cord MRI for MS diagnosis should follow standardised protocols, with current guidance provided by the MAGNIMS (Magnetic Resonance Imaging in MS) consensus guidelines and the Consortium of MS Centers (CMSC) MRI protocols.

Brain MRI

Recommended sequences for brain imaging in the diagnostic workup include:

• FLAIR (fluid-attenuated inversion recovery): The primary sequence for white matter lesion detection. FLAIR suppresses CSF signal, rendering periventricular and juxtacortical lesions conspicuous on a dark background. 3D FLAIR at 3 Tesla is preferred over 2D acquisitions for coverage, resolution, and reformatting capability.
• T2-weighted: Complementary to FLAIR, particularly for posterior fossa lesions where FLAIR is susceptible to flow artefacts and for spinal cord imaging.
• T1-weighted with gadolinium: Essential for identifying actively enhancing lesions (blood-brain barrier disruption), which reflect acute inflammatory activity within the previous 4 to 6 weeks. Standard gadolinium doses (0.1 mmol/kg) administered 5 to 10 minutes before T1 acquisition are standard. Macrocyclic gadolinium agents are preferred over linear agents given the regulatory guidance on gadolinium tissue deposition.
• Pre-contrast T1 (T1 hypointense lesions): Persistently T1-hypointense lesions ("black holes") indicate regions of severe axonal loss and are a marker of irreversible tissue damage. Not required for McDonald Criteria application but relevant for prognosis and treatment decisions.
• DWI (diffusion-weighted imaging): Useful for excluding acute ischaemia mimicking MS relapse, particularly in older patients.
• Advanced cortical sequences (DIR, PSIR): Double inversion recovery and phase-sensitive inversion recovery sequences have substantially higher sensitivity for intracortical and leukocortical lesions than standard FLAIR. Their inclusion in diagnostic protocols improves the sensitivity of cortical lesion detection for DIS, though they are not universally available.

Spinal Cord MRI

Spinal cord imaging should include sagittal and axial T2-weighted sequences with sufficient spatial resolution to distinguish MS lesions from central canal artefact and spondylotic change. Gadolinium-enhanced T1 sequences of the spinal cord are indicated when active spinal disease is suspected. Key radiological characteristics of MS spinal cord lesions that distinguish them from other diagnoses include:

• Length less than two vertebral segments (short-segment myelitis)
• Peripheral location in the cord cross-section (posterior or lateral)
• Affecting less than 50% of the cord cross-sectional area on axial images
• Absence of cord expansion or swelling in chronic lesions
• Cervical and thoracic cord predilection over lumbar cord

Defining a Clinical Attack

The McDonald Criteria define a clinical attack (also called a relapse or exacerbation) as a patient-reported or objectively observed event typical of an acute CNS inflammatory demyelinating episode, either current or historical, with the following characteristics:

• Duration of at least 24 hours
• Occurring in the absence of fever or infection (to exclude Uhthoff’s phenomenon, which is a reversible pseudo-relapse triggered by elevated body temperature)
• Separated from a prior attack by at least 30 days to be counted as a distinct episode
• Ideally corroborated by contemporaneous neurological examination, though historical events with typical symptom patterns (described by the patient) can provide additional contextual evidence at the clinician’s discretion

Common CIS presentations include: unilateral optic neuritis (subacute painful visual loss with relative afferent pupillary defect), partial transverse myelitis (partial sensory or motor disturbance with a cervical or thoracic onset), internuclear ophthalmoplegia (adduction paresis with contralateral nystagmus indicating a medial longitudinal fasciculus lesion), hemispheric syndromes (unilateral weakness, sensory loss, or visual field defect), and brainstem or cerebellar syndromes (diplopia, vertigo, ataxia, dysarthria).

Certain symptom characteristics increase the specificity of an attack for MS over other diagnoses: Lhermitte’s sign (electrical shooting sensation down the spine on neck flexion, indicating cervical cord demyelination), trigeminal neuralgia in a young patient, paroxysmal tonic spasms, and heat sensitivity (Uhthoff’s phenomenon). Conversely, features that reduce specificity and should prompt investigation for alternative diagnoses include: simultaneous bilateral visual loss, longitudinally extensive spinal cord lesions (more than three vertebral levels), severe encephalopathy at onset, and nausea or hiccups suggesting area postrema involvement.

The “No Better Explanation” Requirement

The McDonald Criteria explicitly require that MS is the most plausible explanation for the clinical and radiological findings, with no alternative diagnosis better accounting for the presentation. This principle underlies the enduring observation that MS is the most commonly misdiagnosed neurological condition, with studies suggesting that 10 to 25% of patients carrying an MS diagnosis in tertiary neurology centres may have an alternative aetiology.

MS misdiagnosis causes substantial harm: patients receive disease-modifying immunomodulatory therapy for a condition they do not have, while the true underlying diagnosis is delayed. Common causes of MS misdiagnosis include:

• Neuromyelitis optica spectrum disorder (NMOSD): AQP4-IgG positive NMOSD causes severe optic neuritis, longitudinally extensive transverse myelitis, area postrema syndrome, and other brainstem attacks. MRI may show periventricular lesions (occasionally) but characteristically lacks the extensive juxtacortical, cortical, and infratentorial lesion burden of MS. Anti-AQP4 antibody testing (cell-based assay) should be performed in all patients with atypical CIS features, especially bilateral optic neuritis, longitudinally extensive myelitis, or area postrema involvement.
• MOG antibody-associated disease (MOGAD): Anti-MOG IgG produces optic neuritis (often bilateral or recurrent), cortical encephalitis with seizures, ADEM-like presentations, and myelitis. MOGAD may closely mimic MS on MRI. Anti-MOG antibody testing (cell-based assay) is indicated in patients with ADEM-like presentation, bilateral simultaneous optic neuritis, or optic neuritis with perineural enhancement.
• CNS vasculitis: Primary CNS vasculitis and secondary vasculitis (e.g. SLE, Sjögren’s syndrome) may produce multifocal white matter lesions mimicking MS. Vessel wall enhancement, leptomeningeal enhancement, and stroke-like lesion distribution suggest vasculitis. Serology (ANA, anti-dsDNA, anti-SSA/SSB, ANCA) and brain MR angiography or conventional angiography may be required.
• Small vessel cerebrovascular disease: Non-specific periventricular and subcortical white matter lesions from hypertensive small vessel disease or CADASIL may superficially resemble MS lesions. MS lesions are more likely periventricular with a Dawson’s fingers pattern, cortical or juxtacortical, and infratentorial; vascular lesions more commonly affect the basal ganglia and subcortical U-fibres are typically spared in MS.
• Sarcoidosis: CNS sarcoidosis produces cranial nerve palsies, leptomeningeal enhancement, hypothalamic lesions, and intraparenchymal enhancing masses. Serum and CSF ACE, chest CT, and skin or lymph node biopsy help clarify the diagnosis.
• Susac syndrome: A rare immune-mediated endotheliopathy causing the triad of encephalopathy, branch retinal artery occlusions, and sensorineural hearing loss. Corpus callosum “snowball” lesions (central, rounded lesions in the body of the corpus callosum) on MRI are characteristic.
• Migraine with aura: Subcortical and periventricular white matter lesions occur in patients with migraine and may be misinterpreted as CIS, particularly in young women. Migraine lesions tend to be non-specific, small, and lacking the Dawson’s fingers distribution of MS.
• Lyme neuroborreliosis: Lyme disease affecting the CNS may produce white matter lesions, cranial nerve palsies (particularly facial nerve), and radiculopathy. Borrelia serology (ELISA with Western blot confirmation) is appropriate in endemic areas or patients with exposure history.
• Metabolic myelopathies: B12 deficiency myelopathy, copper deficiency myelopathy, and HIV vacuolar myelopathy may mimic PPMS. Serum B12, methylmalonic acid, homocysteine, serum copper and caeruloplasmin, and HIV serology should be checked in progressive myelopathy presentations.

Paediatric MS and the McDonald Criteria

The 2017 McDonald Criteria are validated primarily in adults, and their application in children requires additional caution. Paediatric MS accounts for approximately 3 to 5% of all MS cases, with 97% of paediatric MS having an RRMS course. Key differences in the diagnostic evaluation of children include:

• ADEM-like first presentations: Children under 11 years of age presenting with polyfocal demyelination and encephalopathy often have ADEM (acute disseminated encephalomyelitis), a monophasic inflammatory demyelination distinct from MS in its aetiology, natural history, and treatment. The 2017 McDonald Criteria cannot be applied to a first episode meeting ADEM criteria; MS can only be considered after a second non-ADEM episode.
• Higher incidence of MOGAD: Anti-MOG antibody disease is proportionally more common in children than adults and produces ADEM, optic neuritis, and myelitis that can be confused with MS. MOG antibody testing should be considered in paediatric patients before an MS diagnosis is made.
• Modified IPMSSG criteria: The International Paediatric Multiple Sclerosis Study Group (IPMSSG) has published specific criteria for paediatric MS diagnosis that incorporate the McDonald framework with ADEM-specific exclusions.
• Impact of disease-modifying therapy: FDA-approved and EMA-approved disease-modifying therapies for paediatric MS now include interferon-β, glatiramer acetate, fingolimod, and dimethyl fumarate. Early and accurate diagnosis is critical to timely access to these agents.

Diagnostic Performance of the 2017 Criteria

Validation studies comparing the 2017 McDonald Criteria to the 2010 criteria in CIS cohorts have consistently demonstrated:

• Increased sensitivity: Approximately 13 to 20% more CIS patients receive an immediate MS diagnosis under 2017 criteria compared to 2010, primarily due to the OCB substitution for DIT. This improvement is most pronounced in centres with high rates of CSF analysis in CIS workup.
• Maintained specificity: Approximately 95 to 98% specificity in validation cohorts when criteria are applied to patients with typical CIS and after appropriate exclusion of NMOSD, MOGAD, and other mimics. The specificity of the OCB-based diagnosis is dependent on confirmatory DIS by MRI; OCBs without DIS should not be used to diagnose MS.
• Positive predictive value for CIS-to-MS conversion: OCB-positive CIS patients have a conversion risk to clinically definite MS of 60 to 80% at 10 years compared with 20 to 30% in OCB-negative patients. This differential underpins the decision to include OCBs as a DIT substitute.
• Performance in diverse populations: The criteria have been validated in European, North American, and Asian cohorts with generally consistent results, though sensitivity may be modestly lower in Asian populations where MS prevalence is lower and the differential diagnosis of NMOSD is proportionally higher.

Impact on Treatment Decision-Making

The primary clinical motivation for updating the McDonald Criteria is to enable earlier initiation of disease-modifying therapy (DMT). Evidence from clinical trials of early intervention in CIS (including BENEFIT, PreCISe, and REFLEX trials) demonstrates that starting DMT at the CIS stage reduces the risk of a second clinical attack and conversion to CDMS, delays disability accrual, and reduces lesion accumulation on MRI compared with delaying treatment until a second attack.

The 2017 criteria therefore have direct therapeutic implications. A patient who meets MS criteria at first presentation is eligible for immediate DMT initiation under the guidelines of most major neurology societies (EAN, AAN, ECTRIMS). By contrast, a patient who has a CIS but does not yet meet McDonald Criteria is classified as having CIS rather than MS and has a narrower range of approved treatment options (interferon-β and glatiramer acetate are approved for CIS at high risk of conversion in some jurisdictions, while higher-efficacy therapies typically require an MS diagnosis).

The decision about which DMT to use after an MS diagnosis is made is guided by disease activity and severity, patient risk tolerance, reproductive plans, comorbidities, and treatment access rather than by the McDonald staging itself. Nonetheless, the McDonald Criteria-determined MS diagnosis is the gatekeeping step for DMT access in most healthcare systems.

Limitations of the McDonald Criteria

• Typical CIS populations only: The criteria are validated for and should be applied exclusively to patients with typical CIS presentations (optic neuritis, partial myelitis, brainstem syndrome, hemispheric syndrome). Applying them to atypical presentations, encephalopathic illness, bilateral optic neuritis, or longitudinally extensive myelitis without first ruling out NMOSD and MOGAD significantly increases the risk of misdiagnosis.
• Requirement for strict exclusion of alternatives: The "no better explanation" principle requires active, systematic exclusion of MS mimics. The criteria do not specify a required minimum diagnostic panel for exclusion, leaving this to clinical judgement. Variations in the thoroughness of differential diagnosis workup between clinicians and centres contribute to diagnostic variability and the ongoing problem of MS misdiagnosis.
• MRI protocol dependency: The criteria assume MRI of adequate quality acquired on scanners of at least 1.5 Tesla using standardised sequences and gadolinium contrast. Non-standardised MRI (e.g. obtained on low-field scanners, without gadolinium, or on inadequate sequences) may fail to demonstrate lesions that would meet DIS or DIT criteria, leading to underdiagnosis, or may produce artefacts mistaken for lesions.
• OCB testing variability: CSF OCB analysis by isoelectric focusing requires specific laboratory expertise and standardised methodology. Results may vary between laboratories, and some published series using nephelometry or ELISA for IgG detection rather than IEF have poorer sensitivity for OCBs. The criteria imply IEF-based OCB detection with paired serum analysis.
• Limitation to RRMS and PPMS: The criteria do not formally address secondary progressive MS, which is typically a clinical diagnosis based on demonstrated transition from RRMS. Nor do they apply to progressive forms of other inflammatory CNS diseases that may superficially meet the criteria’s progression arm.
• No biomarker weighting or risk scoring: The McDonald Criteria are binary (MS or not MS, or additional data needed) rather than probabilistic. They do not quantify the likelihood of MS or incorporate prognostic biomarkers (NfL, CHI3L1, serum NfL) that are increasingly used clinically.
• Not applicable to paediatric ADEM: As noted above, the 2017 criteria cannot be applied to a first demyelinating episode meeting ADEM criteria in children.
• Treatment-era mismatch for progression criteria: The PPMS criteria require one year of objective progression, which may delay diagnosis and treatment in a disease where early ocrelizumab initiation has been shown to reduce disability accumulation. The minimum one-year threshold balances specificity against access to treatment but may not reflect optimal clinical care in all patients.

How to Use This Calculator

The CalcMD McDonald Criteria calculator guides the clinician through the 2017 algorithmic criteria in a structured step-by-step format. To use it accurately:

1. Select the clinical presentation: Choose the option that best describes the patient’s clinical situation. The five options correspond to the five rows of the McDonald Criteria table: (1) two or more attacks with two or more objective lesions, (2) two or more attacks with one objective lesion, (3) one attack with two or more objective lesions, (4) one attack (CIS) with one objective lesion, and (5) insidious progressive course suggestive of PPMS.
2. Enter MRI DIS status: Indicate whether MRI demonstrates T2 lesions in two or more of the five characteristic CNS areas (periventricular, cortical or juxtacortical, infratentorial, spinal cord, optic nerve). Both symptomatic and asymptomatic lesions count. Only one lesion per qualifying area is required; there is no minimum lesion count per area beyond one.
3. Enter MRI DIT status: Indicate whether MRI demonstrates simultaneous enhancing and non-enhancing lesions, or a new T2 or enhancing lesion on follow-up MRI. Do not mark DIT positive for CSF OCBs here; OCBs are entered separately.
4. Enter CSF oligoclonal band status: Indicate whether isoelectric focusing of paired CSF and serum demonstrates two or more CSF-specific IgG bands (absent in serum). This field serves as the DIT substitute in the 2017 criteria for presentations 3 and 4, and as one of the three supportive criteria for PPMS.
5. For progressive presentations (PPMS): Additionally indicate whether one year of objective disability progression independent of relapses is documented, whether two or more qualifying brain T2 lesions are present, and whether two or more spinal cord T2 lesions are present.

The calculator returns one of three outputs: MS Diagnosis Can Be Made (all required criteria are met), Additional Data Needed (current information is insufficient and specific missing data are itemised), or (for the progressive presentation) progressive MS criteria assessment. The output specifies which criteria are fulfilled and which remain outstanding.

The calculator is intended as an educational and clinical decision support tool. A diagnosis of MS is a significant, life-altering diagnosis with major therapeutic, occupational, and psychological implications. It must be made by a qualified neurologist or MS specialist with full access to the patient’s clinical history, neurological examination, neuroimaging, and laboratory results, and with active exclusion of alternative diagnoses, in accordance with the 2017 McDonald Criteria and applicable clinical guidelines.