FOSUN HEALTH Logo

Epilepsy Diagnosis

catalogue

Epilepsy Diagnosis

Epilepsy Diagnosis

Epilepsy is diagnosed from a description. Everything else — the EEG, the MRI, the blood tests — is there to classify the epilepsy, to find a cause, and to decide what to do about it, and none of them can substitute for knowing what actually happened. That is why the first appointment is mostly talking, and why the most valuable thing a patient can bring is not a scan but a person who watched the event, or a phone that recorded it.

First question: are these events seizures?

The question is asked before any other, and it is asked because a substantial share of events labelled as seizures are not. The features that carry the most weight are the stereotyped nature of the episode, its duration, what happened at the very start, and what the person was like afterwards. A witnessed account of the start, the middle and the recovery is worth more than a normal scan, and a video recorded on a phone during an event is worth more than both.

The question is not answered by ruling everything else out. A diagnosis is made positively: from the description of a seizure, or from a recorded event on video EEG with the electrical activity during it. Where the events turn out not to be seizures, the correct next step is to say what they are — syncope, a functional non-epileptic attack, a sleep parasomnia, a movement disorder —

The history, and why it carries the diagnosis

The parts that matter are specific. What was happening in the seconds before — a rising feeling, a smell, a sound, a fear, a tingling. What an observer saw — the eyes and the head, one side or both, stiffening then jerking or jerking alone, the colour of the face, whether breathing stopped. How long, and how the ending came. What the person was like afterwards, and for how long. Whether every episode has been the same.

The rest of the history is about cause and about classification. Birth history and early development, febrile seizures, head injuries with loss of consciousness, meningitis or encephalitis, stroke, and surgery on the brain. Family history of epilepsy or of febrile seizures. Every antiseizure drug ever taken, with the dose reached, how long it was held and why it stopped — because a drug that was abandoned at a low dose after a rash is not the same thing as a drug that failed at the maximum tolerated dose.

Examination is usually normal, and that is expected rather than reassuring enough to stop. What is looked for are the things that point to a cause: asymmetrical findings, a skin marker of a neurocutaneous syndrome , and cognitive or developmental findings that place the epilepsy in a broader syndrome.

EEG: what it can and cannot say

An EEG records the electrical activity of the brain from the scalp over minutes to hours. Its value is in classification rather than diagnosis: epileptiform activity between seizures supports the diagnosis and, critically, distinguishes a generalised epilepsy from a focal one, which is the distinction that chooses the drug. A generalised epilepsy has a characteristic bisynchronous pattern; a focal one has spikes over the region where seizures begin.

What it cannot do is exclude epilepsy. The brain misbehaves occasionally and the recording samples a short window, so a normal EEG is common in people who unquestionably have epilepsy, and repeating it, recording it after sleep deprivation, or recording it for longer increases what is found. The other half of the caution is the reverse: epileptiform activity occurs in a small proportion of people who have never had a seizure, so an abnormal EEG in someone with events that do not look like seizures is a reason for monitoring, not a diagnosis.

MRI: looking for the cause

Imaging in epilepsy is not looking for a tumour in general; it is looking for the specific lesions that cause focal epilepsy, and it is read differently for that reason. Hippocampal sclerosis — scarring and volume loss in the memory structure on one side — is the commonest finding in drug-resistant temporal lobe epilepsy and is the lesion surgery is most often aimed at.

The practical point is one of reporting. A scan acquired and reported through a general pathway can describe the same brain as normal or as showing non-specific changes, while the same images read by someone looking for hippocampal volume loss, a blurred grey-white boundary or a bottom-of-sulcus dysplasia give an answer. Where a report says normal and the clinical picture is focal, the images are worth a second reading rather than the conclusion that there is no cause.

Video monitoring, genetic and immune testing, and functional imaging

Long-term video EEG monitoring records the person continuously, with the EEG, for days, with the aim of capturing a typical event. It is used when the diagnosis is uncertain, when the events are frequent and disabling, when surgery is being considered, and when functional non-epileptic attacks are a possibility that has to be resolved rather than suspected. It is the study that settles mimics, and in pre-surgical evaluation it is what localises the seizure onset.

Genetic testing is considered where the picture suggests a genetic epilepsy, where the result would change the drug chosen, or where it would answer a question the family has asked about other children. Metabolic testing has a specific place: a small number of metabolic diseases present with seizures and have treatments that are not antiseizure medication. Antibody testing is requested where the course suggests autoimmune encephalitis — a subacute onset, memory change, psychiatric features, or seizures that do not respond — because that diagnosis moves treatment from antiseizure drugs to immunotherapy, and the two are not interchangeable.

PET and ictal SPECT belong to the pre-surgical evaluation rather than to the initial workup.

Naming the syndrome, and why it changes the answer

Once the seizure type is established, the syndrome is named from the seizure types present, the age at onset, the EEG pattern, the imaging and the course. This is not an academic exercise: the syndrome carries the outlook, the expected response to medication, and how long treatment is likely to continue, while the seizure type carries only the drug. Two people with the same tonic-clonic seizure can be given very different answers about the future.

Juvenile myoclonic epilepsy responds well to medication but is typically lifelong, is strongly provoked by sleep deprivation and missed doses, and needs a specific discussion about valproate in women of childbearing age because of the risk to a fetus, which is a genuine trade-off rather than a footnote. Temporal lobe epilepsy with hippocampal sclerosis often resists medication and is the condition in which surgery has the strongest case.

What a complete workup produces

A workup worth having answers six questions, and a referral that cannot answer all six is usually worth completing before a decision is made. Are the events seizures, and on what evidence. Which seizure type, and does the seizure begin in one region or across both hemispheres. Is there a structural, genetic, metabolic or immune cause, and has the imaging been read with epilepsy in mind. What does the EEG show, and was it recorded after sleep deprivation or for long enough to be informative. Which drugs have been tried, at what dose, for how long, and why each stopped. And what the seizures currently prevent — driving, work, bathing, sleep, study.

From those six, a plan follows in the same order: whether medication is started now or after a second event; which drug, chosen for the seizure type and for the person rather than for the label; whether a cause needs treating in its own right; and, where two appropriate drugs have already failed, whether drug resistance is established and surgery or stimulation should be assessed. The last of those is the step most often delayed, and it is the one where an opinion from another centre is most often worth having.

Frequently Asked Questions

How many seizures does it take to be diagnosed with epilepsy?

The usual answer is two unprovoked seizures more than 24 hours apart. Epilepsy can also be diagnosed after a single seizure where the risk of another is high enough — because of an abnormality on imaging, an unequivocal epileptiform EEG, or a recognised syndrome — and in that situation starting treatment is discussed rather than assumed. The definition is about the chance of recurrence, not about a count that has to be reached.

Why does my doctor keep asking what happened before I lost consciousness?

Because the beginning of the seizure is the part that says where it started. A rising feeling in the stomach, a smell, a sense of dread or a tingling in one hand are all focal seizures that spread, and they place the onset in a specific lobe. That is what separates a focal epilepsy from a generalised one, and the distinction decides which medication is used, because some drugs suit focal epilepsy and a few make generalised seizure types worse.

Do I need a scan if my seizures started years ago and I am well controlled?

Usually the answer is that the imaging should have been done once, and if it was done with an epilepsy protocol and reported, it does not need repeating. Where it is worth revisiting is if seizures change character, if control is lost after years of stability, if new findings appear on examination, or if the original scan was a CT, which misses the lesions that matter most in focal epilepsy.

What happens during video EEG monitoring?

You stay on a ward for several days with EEG electrodes attached and continuous video and EEG recording, and antiseizure medication is usually reduced to allow a typical event to occur, which is done deliberately and with monitoring rather than as a risk. There is a button to press when an event begins, and a member of staff comes. If no event occurs, the admission still yields information about how often events are happening and what the EEG shows between them.

Send the reports and the drug history

Send a description of an event from someone who saw it, or a recording of one, together with every EEG report and whether it was recorded after sleep deprivation, the imaging reports and whether the MRI was read with an epilepsy protocol, and the list of antiseizure drugs with the dose reached, how long each was taken and why it stopped. A neurologist with an epilepsy interest will return a written opinion within about 48 hours of complete records arriving: whether the events are seizures and which type, whether the workup is complete, what one investigation would settle the remaining question, and whether drug resistance is genuinely established.

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*