How MS is diagnosed
Last reviewed November 16, 2026. OwieClear editorial team. Waiting for clinician review. Not medical advice. Talk with your doctor about your own care.
MS is diagnosed by assembling evidence, not by one result. The requirement is damage scattered in space, meaning more than one part of the central nervous system, and scattered in time, meaning it did not all happen at once.
Just as important, and easy to overlook, is ruling out the conditions that mimic it. Misdiagnosis is well documented, and some of the mimics need completely different treatment.
Related pages: Multiple sclerosis (MS)
The bottom line
- Diagnosis follows the McDonald criteria, which combine clinical history, MRI and spinal fluid findings.
- MRI is the central investigation.
- Oligoclonal bands in the spinal fluid support the diagnosis and can substitute for evidence of damage over time.
- Several other conditions must be excluded first.
- How often MS is diagnosed in people who do not have it, with published estimates varying.
- How best to diagnose MS in older adults, where unrelated white matter changes are common.
The story, step by step
- 1. The story ()
- 2. MRI ()
- 3. Spinal fluid ()
- 4. Other tests ()
- 5. Ruling out ()
What the evidence labels mean
History and examination
What happens
The pattern of symptoms over time carries as much weight as any scan.
What the research shows
- A neurologist asks about episodes of neurological symptoms, including ones years earlier that were dismissed at the time: a few weeks of blurred vision in one eye, numbness up one side, sudden double vision.
- Symptoms that came on over hours to days, lasted weeks, then improved are the typical shape of a relapse.
- Examination looks for signs the person may not have noticed: altered reflexes, changed eye movements, subtle weakness, reduced vibration sense, or a pale optic disc from past optic neuritis.
- Specific signs carry weight: internuclear ophthalmoplegia, where the eyes fail to move together, and Lhermitte's sign, an electric sensation down the spine on bending the neck.
- Symptoms that are steady and unvarying, or purely subjective with a normal examination, point away from MS.
Why it matters
Old episodes matter. It is worth writing down anything unexplained that lasted days to weeks, even years ago.
What's still uncertain
- Recall of past episodes is unreliable, and some people have none to report.
Read more
- NINDS: Multiple SclerosisGovernment
- National Multiple Sclerosis SocietyNonprofit
Next: MRI
The main test
What happens
Magnetic resonance imaging shows the lesions and, crucially, whether they happened at different times.
What the research shows
- Brain and spinal cord are both scanned. MS lesions are typically oval, sit beside the ventricles pointing outward in a pattern called Dawson's fingers, and also appear next to the cortex, in the brainstem and cerebellum, and in the spinal cord.
- Dissemination in space requires lesions in at least two of four typical regions.
- Dissemination in time can be shown by lesions of clearly different ages on a single scan: one taking up gadolinium contrast, meaning the barrier is currently leaking there, alongside one that is not.
- Otherwise it is shown by a new lesion appearing on a later scan.
- Spinal cord lesions in MS are typically short, affecting fewer than three vertebral segments, and partial across the cord. A long lesion spanning several segments points toward a different diagnosis.
Why it matters
The scan does two jobs at once: it shows where the damage is and, through contrast, roughly when it happened.
What's still uncertain
- White matter spots are common with age, migraine and vascular disease, and can be mistaken for MS.
Read more
- NINDS: Multiple SclerosisGovernment
- National Multiple Sclerosis SocietyNonprofit
Next: Spinal fluid
The lumbar puncture
What happens
A sample of cerebrospinal fluid shows whether antibodies are being made inside the nervous system.
What the research shows
- A needle is placed into the lower back below the level of the spinal cord and a small amount of fluid is removed.
- The key finding is oligoclonal bands: antibody bands present in the fluid but not in a blood sample taken at the same time, meaning they are produced inside the central nervous system. They are found in the large majority of people with MS.
- Under current criteria, oligoclonal bands can stand in for evidence of damage over time, allowing diagnosis after a single episode.
- Kappa free light chain measurement is an automated alternative being adopted in some centres.
- The fluid also helps exclude infection and other inflammatory conditions, which is part of why it is done.
Why it matters
Bands are not unique to MS, so they support a diagnosis rather than making one. Headache afterwards is the common side effect, and it is treatable.
What's still uncertain
- A minority of people with MS do not have oligoclonal bands, so a negative result does not exclude it.
Read more
- NINDS: Multiple SclerosisGovernment
- MedlinePlus: Multiple SclerosisGovernment
Next: Other tests
Evoked potentials and bloods
What happens
Supporting tests measure conduction speed and exclude alternatives.
What the research shows
- Visual evoked potentials measure how long a signal takes to travel from the eye to the back of the brain. Delay suggests past optic nerve demyelination even where vision has recovered.
- Blood tests are done mainly to exclude mimics: vitamin B12, thyroid function, inflammatory markers, antinuclear antibodies, and tests for infections including syphilis, HIV and Lyme disease where relevant.
- Aquaporin-4 antibodies test for neuromyelitis optica spectrum disorder, and MOG antibodies for MOGAD. Both can look like MS and both are treated differently, with some MS drugs making NMOSD worse.
- Serum neurofilament light chain, a marker of nerve damage, is used in research and increasingly in monitoring, but it is not diagnostic on its own.
- Optical coherence tomography scans the retina and can show thinning after optic neuritis.
Why it matters
The antibody tests matter: treating NMOSD as if it were MS can cause harm, so they are routinely checked when the picture is atypical.
What's still uncertain
- Neurofilament light is promising but not yet standardized for individual decisions.
Read more
- NINDS: Multiple SclerosisGovernment
- OwieClear: Neuromyelitis optica (NMOSD or MOGAD)OwieClear
Next: Ruling out
What else it could be
What happens
The criteria require that no better explanation exists, and that step is not a formality.
What the research shows
- Vitamin B12 deficiency, which damages the spinal cord and is treatable.
- NMOSD and MOGAD, which cause optic neuritis and spinal cord inflammation and have their own treatments.
- Neurosarcoidosis, central nervous system vasculitis, lupus and Sjogren's syndrome.
- Infections including Lyme disease, syphilis and HIV.
- Small vessel disease and migraine, both of which produce white matter spots on MRI, particularly in people over fifty.
- Inherited conditions such as leukodystrophies and CADASIL, and compressive problems in the neck that mimic a spinal cord lesion.
Why it matters
A diagnosis of MS should rest on a pattern, not on a scan report alone. A second opinion is reasonable where the picture is atypical.
What's still uncertain
- Red flags for an alternative diagnosis include a steadily worsening course without relapses in a young person, prominent headache, fever, or normal spinal fluid with an unusual MRI.
Read more
- NINDS: Multiple SclerosisGovernment
- National Multiple Sclerosis SocietyNonprofit
More plain-language reading
- OwieClear: The four patterns of MSOwieClear
- OwieClear: MS symptoms in detailOwieClear
- OwieClear: What happens in demyelinationOwieClear