FOSUN HEALTH Logo

Multiple Sclerosis (MS) Diagnosis

catalogue

Multiple Sclerosis (MS) Diagnosis

Multiple Sclerosis (MS) Diagnosis

There is no single test that comes back positive for this condition, which is the first thing worth saying because it is the source of most of the confusion. What the diagnosis requires is two demonstrations : that the damage is in more than one place within the central nervous system, and that it happened on more than one occasion. Both can come from a scan, both can come from the history, and one can come from the spinal fluid in the right context — but both have to come from somewhere, and 'the scan showed some lesions' is neither. Alongside that, there is a second job that is just as important and is done far less often: excluding the conditions that look like this one , of which two have different treatments and different prognoses. This page sets out what has to be shown, what each test establishes and what it does not, why a scan alone is not a diagnosis, and what a finished assessment should leave behind in writing.

The question that has to be answered twice

Before any of the machinery is brought to bear, there are two questions, and they are sequential rather than parallel. The first is whether the damage is in more than one region of the central nervous system — called dissemination in space. The second is whether it happened on more than one occasion — dissemination in time. A diagnosis needs both. Neither is a formality; together they are the entire difference between this condition and the much commoner finding of a few bright spots on a scan.

Dissemination in space is demonstrated by lesions in more than one of the four regions this disease favours: the white matter around the ventricles, the cortex or the area just beneath it, the infratentorial compartment — brainstem and cerebellum — and the spinal cord. It can also be demonstrated clinically, by symptoms that belong to two different pathways at the same time, and historically it was demonstrated by evoked potentials showing slowing in a pathway the person had no symptoms in.

Dissemination in time is the harder of the two, and the one most often simply assumed. It is demonstrated in one of three ways: a new lesion on a follow-up scan compared against the earlier one; an enhancing and a non-enhancing lesion together on a single scan, which means one is active now and one happened earlier; or, in the right clinical setting, the spinal fluid pattern that goes with this disease. What it is not demonstrated by is a single scan showing several lesions of the same age, which is compatible with one event having caused all of them.

The reason this matters practically is that the answer decides whether a person is started on long-term treatment. A first episode in someone whose scan shows lesions in one region and no evidence of anything older is a clinically isolated syndrome — a state in which treatment may still be appropriate, but in which the diagnosis has not been made. Saying so plainly, rather than writing the name down early, is what allows the second opinion later to be honest about what was and was not shown.

What the MRI establishes

Magnetic resonance imaging is the study that carries most of the diagnostic weight here, and it is worth being precise about which of the two questions it answers and which it does not. It is also worth being precise about what has to be imaged, because the commonest technical failure in this condition is a scan of the wrong part.

Finding What it establishes What it does not settle
Lesions in more than one characteristic region Dissemination in space, which is the first of the two requirements Not that they are caused by this disease. Their shape and position are the argument, not their number
A lesion that enhances with contrast That the blood-brain barrier is breached there now — activity at this moment rather than at some point Not how the person feels. Active lesions are frequently silent, and disabling symptoms frequently come from lesions that are not enhancing
Enhancing and non-enhancing lesions together Dissemination in time on a single scan — the strongest single piece of evidence available Nothing about the future. It dates the disease; it does not stage it
A new lesion against a previous scan Dissemination in time, and — where the person has no symptoms — activity that would otherwise have been invisible Not a reason to change treatment on its own. It is one input into a decision that also weighs the clinical picture
Cord lesions Often the missing region that completes dissemination in space, and the most common omission when only the brain is imaged Not specific to this disease. A long cord lesion spanning several segments points away from it and towards the neuromyelitis optica spectrum
The total lesion count A baseline against which future scans are compared, and nothing more Not a score that tracks disability. The correlation between what a scan shows and what a person can do is poor, and reading the two together is one of the most reliable ways to get both wrong
A normal scan That nothing was seen — on that scanner, with that protocol, in the regions imaged Not that the disease is absent. Where the symptoms point to the cord and only the brain was imaged, a normal scan is an unanswered question

Two practical points complete it. The first is that the scan has to include the cord where there is any symptom below the neck, and it has to be done with contrast where the question of current activity is open — because without contrast the single most informative finding on the list above is not available. The second is that the comparison matters: a scan reported without the previous one beside it cannot say whether anything is new, and in this condition 'new' is the word that decides.

What the spinal fluid adds

Lumbar puncture has two separate jobs in this assessment, and they are worth separating because people are usually told only about the first. The first is supportive and the second is exclusion — and the second is the one that changes management.

Supportive. The finding that matters is oligoclonal bands present in the spinal fluid but not matched in the blood, which reflects a local immune response within the central nervous system rather than a systemic one. Where the imaging is not sufficient on its own, this pattern can contribute to demonstrating dissemination in time. Where the imaging already demonstrates both requirements, it adds confirmation rather than information. And where it is absent, it does not exclude the disease — a point worth stating because a negative result is regularly read as closure when it is not.

Exclusion. The fluid is where infection and other inflammatory conditions are found, and several of them produce a picture that a scan alone cannot distinguish from this one. Where the presentation is atypical — progressive from the outset, a long cord lesion, bilateral optic neuritis, or anything with a fever attached — this is not an optional step. It is the test that answers the question that is actually open.

What is taken is worth knowing, because the extra tubes cannot be added afterwards: opening pressure, cell count and differential, protein, glucose, oligoclonal bands paired against a blood sample taken at the same time, and — where the question calls for it — material held for microbiology and cytology. Two things are asked for that are regularly forgotten: that the paired blood sample is drawn, because bands without the comparison are uninterpretable; and that the date of the puncture is recorded relative to any treatment, because that changes how everything is read.

What else has to be excluded, by name

This is the half of the assessment that is most often left undone, and it is the half that most often brings people for a second opinion. A file that says 'demyelinating disease' without the tests below is a file in which the differential was considered in principle rather than excluded in fact.

Alternative What points at it Why it matters if it is missed
Neuromyelitis optica spectrum disorder Severe myelitis with a long cord lesion, bilateral or sequential optic neuritis, intractable hiccough and vomiting, and a normal or near-normal brain scan It is a different disease with a different treatment and a different prognosis. Some of the medicines used in MS make it worse
MOG antibody disease Optic neuritis that is often bilateral and often steroid-responsive, sometimes with a myelitis, sometimes in children Also a different disease, and also one where the MS treatment pathway is not the right one
Other autoimmune and inflammatory disease A systemic picture — rash, joints, dry eyes and mouth, a raised inflammatory marker, or involvement outside the nervous system The treatment is directed at the underlying condition rather than at the nervous system alone
Infection Fever, a subacute course, or an immunosuppressed state; and a range of organisms that reach the nervous system Treating it as MS means giving immunosuppression to someone with an infection, which is the error with the shortest fuse
Nutritional and metabolic causes Vitamin B12 deficiency, copper deficiency, and a small number of inherited conditions that affect the cord Some of them are entirely reversible, which is why they are tested rather than assumed away
Vascular disease Age and a vascular risk profile, lesions that follow a vascular territory, and no enhancement The commonest reason a scan is over-read in an older person, and the commonest reason a progressive gait is attributed to MS
Structural compression of the cord Neck pain, a dermatomal rather than a truncal sensory level, and stiffness that has been progressive It is a surgical problem, and it is excluded by imaging the neck rather than by assuming the diagnosis

The two antibody tests at the top of that table are the ones worth asking for by name, and they are the single most useful thing a person can request at a second opinion. Neither is routine in every assessment, both change management, and neither is reliably requested unless someone asks. Alongside them, the blood work that establishes the rest — inflammatory markers, the autoimmune serology, vitamin B12, and the infectious serology where the context calls for it — is requested because each one changes what is done next, not as a panel to be ticked.

Why a scan with lesions is not a diagnosis

White matter lesions are one of the commonest incidental findings in medicine. They are left behind by migraine, by high blood pressure, by diabetes, by smoking, by radiotherapy and chemotherapy, by a range of infections, and by getting older. In someone scanned for a headache or after a minor head injury, the finding of a few T2 bright spots in the white matter is far more often one of those things than it is this disease.

What separates them is not the number but the shape and the address. Lesions that sit against the ventricles, that involve the corpus callosum, that sit in the brainstem or the cerebellum or the cord, that are oriented away from the ventricles rather than along them — these are the features that make a lesion look like this disease. A handful of non-specific dots in the centrum semiovale in someone with hypertension looks like what it almost always is.

And there is a second requirement that a single incidental scan almost never meets: none of it says anything about time. A scan showing several lesions of the same age is compatible with one event having caused all of them. Without a comparison scan, or without a clinical history of a separate episode, that requirement is simply not addressed.

The consequence of getting this wrong is not academic. A person labelled with this disease on the strength of an incidental scan is started on years of monitoring and, frequently, on treatment; is told they have a lifelong neurological condition; and may have the actual question — what caused the headache they were scanned for — left unanswered. The person with a radiologically isolated syndrome, who has the lesions and no symptoms at all, is the hardest case to resist treating and the one where restraint is the right answer. Naming what has been found, saying what it does not yet mean, and arranging what to watch for is a complete and honest response, and it is not the same as doing nothing.

What a finished assessment leaves behind in writing

Five things, and a family is entitled to ask for each of them. They are also the five things most often missing from the letter that follows a scan.

First, whether dissemination has been demonstrated — in space and in time — and on what evidence, stated as a judgement rather than as a restatement of a radiology report. Second, which regions were imaged, so that a reader can tell whether the cord was included and whether contrast was given. Third, which alternatives have been excluded and on what grounds, naming the antibody tests specifically; a differential that was considered but not tested is not an exclusion. Fourth, what would change the answer, including whether a repeat scan at a defined interval, or a repeat of the fluid, is the right next step — and what interval, rather than 'follow-up as clinically indicated'. Fifth, what the follow-up has to contain: how often imaging is repeated and why, which symptoms are asked about at each visit by name, and what finding would prompt a change in treatment.

That last item is the one that changes a life most, and it is the one most often left out. A review built around whether a new lesion has appeared will not surface fatigue, heat sensitivity, slowed thinking, bladder function or mood. A review that lists them as items will, and it is the difference between a person discovering the invisible half at home over years and having it managed from the first visit.

Frequently Asked Questions

Can MS be diagnosed from one MRI?

Sometimes, and it depends on what the scan shows rather than on how many scans there have been. Dissemination in time can be demonstrated on a single scan when an enhancing lesion and a non-enhancing lesion are present together — one is active now, one happened earlier. If the scan shows lesions in only one region, or shows nothing that dates anything, then time has to be demonstrated another way: a follow-up scan that shows a new lesion, or a clinical episode that happened earlier and can be documented. What cannot happen is the commonest version of all — a scan done for a headache, a report that says demyelination cannot be excluded, and a diagnosis recorded on the strength of it.

Why do I need a lumbar puncture if the MRI already shows lesions?

For two different reasons, and the second is the one that matters. The first is supportive: oligoclonal bands in the fluid but not in the blood are a pattern that goes with this disease, and where the imaging is not quite enough they can supply part of what is needed. The second is exclusion: the fluid is where infection, other inflammatory conditions and some of the mimics are found or ruled out, and that job cannot be done by a scan. A lumbar puncture is not an obligatory step in every assessment, and a normal fluid does not exclude the disease — but where the picture is atypical, or where the mimics have not been excluded, it is the test that answers the question that is actually open.

What if my MRI is normal but I have the symptoms?

That happens, and it is not a contradiction. A normal scan does not exclude the condition, particularly if only the brain was imaged and the symptoms point to the cord, or if the scan was done on a lower field strength or reported without contrast. What should happen is that the question is restated rather than closed: which regions were actually imaged, was the cord included, and what is the next test that would answer it. The error to avoid is being discharged with a normal scan and a file that says no further action, when what the scan did was fail to answer rather than answer in the negative.

How do I know the mimics have been excluded rather than just not thought about?

By asking which tests were done rather than which diseases were considered. The two that change management most are the aquaporin-4 and MOG antibody tests, and they can be asked for by name. A file that says 'demyelinating disease' without either result is a file in which a different disease with a different treatment has not been excluded. The same applies to the blood tests for the inflammatory, infectious and nutritional causes that produce the same picture. It is a reasonable question rather than a challenge, and it is the one most worth asking at a second opinion.

Find out whether dissemination has been demonstrated, and what was never asked

Send the imaging with its report — both the brain and the cord where both were done — any spinal fluid result, and the list of antibody tests that were performed. Alongside them, a short account in your own words of each episode in the order it happened, including what recovered and what did not. Our team will return a written opinion on whether dissemination has been shown, which alternatives have been excluded and on what grounds, and what would answer the question that is still open.

Want to see the whole process first? Your full medical journey, step by step →

Book an Appointment Request a Video Consultation

Follow Fosun Health

Apply for an Appointment

Do you have test results*