The McDonald criteria are the standard diagnostic framework neurologists use to confirm multiple sclerosis (MS). First established in 2001 and revised several times since, the criteria center on one core principle: proving that damage to the central nervous system has occurred in more than one location and at more than one point in time. The most recent revision, published in 2024, expanded the criteria to allow earlier and more accurate diagnosis across a wider range of patients.
The Two Core Requirements
An MS diagnosis rests on demonstrating two things: dissemination in space (DIS) and dissemination in time (DIT). In plain terms, DIS means the disease has affected at least two separate areas of the central nervous system. DIT means the disease has been active on more than one occasion, not just a single episode. Both must be present before a neurologist can diagnose MS, and both can be demonstrated through a combination of clinical symptoms and MRI findings.
There’s also a baseline requirement that never changes between revisions: the clinical and imaging findings must have “no better explanation.” The differential diagnosis for MS is broad. Migraine, Lyme disease, vasculitis, and blood vessel changes from aging can all produce brain lesions that look similar on a scan. Neurologists are expected to rule out these alternatives before applying the McDonald criteria, and to periodically reconsider the diagnosis even after it’s been made.
What Dissemination in Space Looks Like
To satisfy DIS, an MRI needs to show at least one characteristic lesion in at least two of four (now five) specific areas of the central nervous system. In the 2017 version of the criteria, those areas were:
- Periventricular: tissue surrounding the fluid-filled ventricles deep in the brain
- Cortical or juxtacortical: the brain’s outer surface or the white matter just beneath it
- Infratentorial: the lower part of the brain, including the brainstem and cerebellum
- Spinal cord: lesions anywhere along the spine
These lesions appear as bright spots on a specific type of MRI scan called a T2-weighted image. A single lesion in two of these regions is enough. The pattern matters more than the total number of lesions because MS tends to leave damage in these characteristic locations, which helps distinguish it from other conditions.
What Dissemination in Time Looks Like
DIT proves that the disease didn’t just strike once. There are several ways to demonstrate this. The most straightforward is a second clinical attack, a new episode of neurological symptoms occurring weeks or months after the first. But waiting for a second attack delays diagnosis and treatment, so MRI provides alternatives.
On a single MRI scan, DIT can be satisfied if the scan shows both a gadolinium-enhancing lesion (one that lights up with contrast dye, indicating active inflammation) and a non-enhancing T2 lesion (indicating older damage). The contrast between “fresh” and “old” lesions on the same scan proves the disease has been active at different times. Alternatively, a follow-up MRI that reveals a new T2 lesion compared to a previous scan also fulfills DIT.
There’s a third option that avoids waiting for a repeat scan entirely. If a spinal fluid test finds oligoclonal bands, which are specific immune proteins present in the cerebrospinal fluid but absent from the blood, this can substitute for MRI-based evidence of DIT. Finding these bands means the immune system has been chronically active within the central nervous system, which strongly supports an MS diagnosis. Most laboratories look for two or more bands in spinal fluid that don’t appear in a matched blood sample.
Diagnosing Primary Progressive MS
Most people with MS experience a relapsing-remitting course, with distinct flare-ups followed by partial or full recovery. But roughly 10 to 15 percent have primary progressive MS (PPMS), where disability accumulates gradually from the start without clear relapses. The McDonald criteria handle this form differently.
A PPMS diagnosis requires at least 12 months of progressive neurological worsening, plus two of the following three findings: at least one T2 lesion in a typical brain region (periventricular, cortical/juxtacortical, or infratentorial), two or more T2 lesions in the spinal cord, or oligoclonal bands in the cerebrospinal fluid. Because PPMS doesn’t produce the distinct attacks that make DIT easy to demonstrate, the criteria lean more heavily on the pattern of steady progression combined with supportive MRI and spinal fluid evidence.
From First Symptoms to Diagnosis
Many people first encounter the McDonald criteria after experiencing what’s called a clinically isolated syndrome (CIS), a single episode of neurological symptoms like optic neuritis, numbness, or weakness that lasts at least 24 hours and is consistent with MS. CIS is not MS by itself, because a single episode doesn’t prove disease activity over time. But if that first episode comes with MRI findings that already satisfy both DIS and DIT, or if oligoclonal bands are found in the spinal fluid alongside qualifying lesions, a full MS diagnosis can be made without waiting months or years for a second attack.
This matters because early treatment consistently leads to better long-term outcomes. Research on CIS patients shows that finding both oligoclonal bands and MRI lesions significantly reduces diagnostic delays. Some biomarkers in spinal fluid can even predict how likely CIS is to progress: studies have found that 40 to 50 percent of patients with certain elevated markers in their spinal fluid convert to MS within the first year.
What Changed in the 2024 Revision
The 2024 revision made several meaningful updates to the criteria. The most notable is the addition of the optic nerve as a fifth anatomical location for DIS. Previously, optic nerve involvement could support a clinical picture of MS but didn’t formally count toward the spatial requirement on imaging. Now, evidence of optic nerve damage, demonstrated through significant asymmetry in the retinal nerve fiber layer measured by optical coherence tomography (a quick, non-invasive eye scan), can help satisfy DIS alongside the original four regions.
The revision also introduced new tools that can strengthen diagnostic confidence. Two MRI features, the central vein sign (a tiny vein visible in the center of a lesion) and paramagnetic rim lesions (a ring of iron-laden immune cells at a lesion’s edge), can now provide supportive evidence for MS when available. These features help distinguish MS lesions from look-alikes caused by migraine or aging. Kappa free light chains in spinal fluid, a faster and cheaper alternative to traditional oligoclonal band testing, were also formally incorporated.
The 2024 criteria also unified the diagnostic approach across relapsing and progressive forms of MS and across age groups, providing guidance for both pediatric and late-life presentations. For patients over 50, where age-related white matter changes can mimic MS lesions, the criteria now offer specific guidance to reduce the risk of misdiagnosis. In certain cases, the revision even allows a diagnosis when someone has radiologically isolated syndrome, meaning characteristic MS lesions found incidentally on a brain scan done for another reason, without any neurological symptoms yet.
Accuracy and Limitations
The McDonald criteria are highly sensitive, meaning they catch nearly everyone who has MS. A study of patients with clinically isolated syndrome found that the 2017 criteria had 100 percent sensitivity. However, specificity was notably low at just 13.8 percent, meaning a significant number of people who met the criteria at the CIS stage did not ultimately develop MS. This imbalance reflects the criteria’s design priority: catching MS early so treatment can begin, even at the cost of some false positives that require reassessment over time.
This is why neurologists treat a McDonald criteria diagnosis as a living assessment rather than a one-time ruling. The “no better explanation” requirement means the diagnosis should be reconsidered if symptoms evolve in unexpected ways, if treatment doesn’t produce expected results, or if new information points toward an alternative condition. More than 60 known causes of optic nerve inflammation alone exist, and conditions like neuromyelitis optica spectrum disorder can closely mimic MS on early presentation.