Drug-Resistant Epilepsy
What it actually means — and why it does not mean there are no treatments left
Most people with epilepsy are treated first with antiseizure medication.
For many, medication provides good seizure control.
But some people continue to have seizures even after appropriate medicines have been tried correctly.
This is known as drug-resistant epilepsy, often shortened to DRE.
Older terms include:
refractory epilepsy
treatment-resistant epilepsy
pharmacoresistant epilepsy
and intractable epilepsy.
The International League Against Epilepsy — ILAE — defines drug-resistant epilepsy as failure of two tolerated, appropriately chosen and appropriately used antiseizure medication schedules, whether used alone or in combination, to achieve sustained seizure freedom.
The definition published by the ILAE in 2010 remains the internationally used standard and was reaffirmed in ILAE educational discussion in 2025. (ilae.org)
The important number is two
Drug-resistant epilepsy should be considered after two appropriate antiseizure medication treatments have failed to produce sustained seizure freedom.
Those medicines might have been used:
one after another as monotherapy
or as part of combination treatment.
But simply having taken two different epilepsy medicines at some point is not enough.
They need to have been:
suitable for the person's seizure type or epilepsy syndrome
taken appropriately
used at an adequate dose and for an adequate duration
and tolerated sufficiently for a meaningful treatment trial.
Epilepsy Action summarises the same principle: drug resistance is suspected when two appropriate epilepsy medicines have failed to stop seizures despite being suitable, tolerated and used at an effective dose. (epilepsy.org.uk)
“I tried two medicines” does not always mean drug resistance
Imagine someone with a generalised epilepsy was initially prescribed a medicine that was actually inappropriate for their seizure type.
That treatment failing would not necessarily prove the epilepsy was drug resistant.
Likewise, if a medicine had to be stopped almost immediately because of a serious adverse effect before anyone could establish whether it controlled seizures, that is different from a genuine failure of seizure control.
The ILAE definition deliberately uses the words:
appropriately chosen
and
appropriately used.
The aim is to identify epilepsy that has genuinely resisted suitable medical treatment — not epilepsy that has simply not yet been treated correctly.
Side effects and treatment failure are not exactly the same thing
A medicine can fail for different reasons.
It may:
not control seizures
partially reduce seizures without stopping them
produce intolerable adverse effects
or never reach an adequate treatment trial.
For the formal ILAE definition of drug resistance, the important issue is failure of adequately used and tolerated treatments to achieve seizure freedom. (ilae.org)
Someone whose medicines cannot be tolerated still deserves specialist review.
But their situation may not represent exactly the same biological problem as seizures persisting despite adequate treatment.
Does a medicine count as successful if seizures become less frequent?
Reducing seizures can be extremely valuable.
A treatment that changes someone from:
several seizures each week
to:
one seizure every few months
may dramatically improve their life.
But that is not the same as sustained seizure freedom.
The ILAE drug-resistance framework uses seizure freedom as the key treatment outcome rather than simply a percentage reduction in seizure frequency.
This reflects the fact that even occasional seizures can still affect:
safety
independence
employment
driving
injury risk
and quality of life. (ilae.org)
What counts as sustained seizure freedom?
The original ILAE drug-resistance framework defined seizure freedom as:
at least 12 months without seizures, or three times the longest interval between seizures before the treatment began — whichever is longer.
For example, if someone previously sometimes went six months between seizures, a few seizure-free months after changing treatment would not yet prove that the medicine had achieved sustained control.
Long enough follow-up is needed to distinguish real treatment success from the person's normal variation in seizure frequency. (ilae.org)
Auras matter when judging seizure freedom
If a person's familiar aura is genuinely a focal seizure, continuing to experience those events means they are not completely seizure-free.
For example:
Before treatment:
focal aura → impaired consciousness → bilateral tonic-clonic seizure
After treatment:
brief focal aura only
The treatment may have produced a major improvement by preventing seizure spread.
But if the aura is itself an epileptic focal seizure, seizure activity is still occurring.
That distinction is important when assessing treatment response.
Related Information Hub page:
Auras and Epilepsy: The Seizure Before the Seizure?
Does one breakthrough seizure mean a medicine has failed?
Not automatically.
The specialist needs to understand why it happened.
For example:
was medication missed?
was the dose recently changed?
was there vomiting?
was another medicine interacting with it?
had the person actually reached an adequate dose?
was this their normal seizure type?
was there a major acute illness or metabolic problem?
A seizure caused by an avoidable interruption to treatment should not necessarily be interpreted in exactly the same way as repeated seizures despite appropriate stable treatment.
Before calling epilepsy drug-resistant, check whether it really is epilepsy
Persistent attacks despite several medicines should sometimes prompt doctors to revisit the diagnosis.
Conditions that can mimic epilepsy include:
syncope
cardiac rhythm disorders
functional seizures
sleep disorders
movement disorders
migraine
and other neurological conditions.
Some people have also been found to have both epilepsy and another type of event.
Simply adding more antiseizure medicines is unlikely to help an event that is not epileptic.
Tertiary assessment can therefore involve reviewing:
the original history
videos
EEG
video telemetry
MRI
and the person's different event types.
What is pseudo-resistance?
Clinicians sometimes use terms such as pseudo-drug resistance or pseudo-refractory epilepsy when seizures appear resistant to treatment but there is another explanation.
Possible reasons can include:
incorrect diagnosis
incorrect seizure classification
inappropriate medicine
inadequate dose
medicine not taken as prescribed
important drug interactions
or inaccurate seizure counting.
This is not about blaming the person.
There may be very practical reasons treatment has not been taken consistently, including:
difficult side effects
memory problems
complicated dosing schedules
prescription problems
swallowing difficulties
or misunderstanding dose changes.
Identifying the reason matters because it may be fixable.
Correct seizure classification matters
Some antiseizure medicines are suitable for particular seizure types but can worsen others.
If the epilepsy has been classified incorrectly, repeated medication failure may partly reflect the wrong treatment strategy.
That is why specialists may revisit:
focal versus generalised epilepsy
the exact seizure types
EEG findings
and epilepsy syndrome
when seizures remain uncontrolled.
Related Information Hub page:
Focal, Generalised, Unknown and Unclassified Seizures — Understanding the 2025 ILAE Classification
Why two medicine failures matter
The probability of achieving lasting seizure freedom generally falls as additional appropriate medicines fail.
That does not mean later medicines can never work.
Some people do obtain seizure control after several treatment changes.
But failure of two appropriate treatments is an important clinical signal that simply continuing an endless sequence of medication trials may not be enough.
The ILAE continues to use this threshold because it identifies a group in whom other treatment strategies should be actively considered. (ilae.org)
How common is drug-resistant epilepsy?
Approximately three in ten people with epilepsy continue to have seizures despite trying appropriate antiseizure medicines.
The ILAE describes roughly one-third of people with epilepsy as continuing to have seizures despite medication, while Epilepsy Action gives a similar figure of around three in ten. (ilae.org) (epilepsy.org.uk)
That makes drug-resistant epilepsy a major part of epilepsy care rather than an exceptionally rare situation.
Drug-resistant does not mean untreatable
This is perhaps the most important message on this page.
Drug-resistant epilepsy does not mean nothing else can be done.
It means that standard antiseizure medication alone has not yet achieved sustained seizure freedom.
Other possibilities may include:
different medicine combinations
epilepsy surgery
vagus nerve stimulation
other neuromodulation approaches
ketogenic dietary therapy in selected cases
syndrome-specific treatments
and treatment directed at an underlying cause.
The correct next step depends on the person's epilepsy.
Referral to a tertiary epilepsy service
A tertiary epilepsy service is a specialist centre able to provide investigations and treatments beyond routine neurology care.
NICE says people should be referred to a tertiary epilepsy service to be seen within 4 weeks when:
diagnosis, cause, seizure type or epilepsy syndrome remains uncertain
the epilepsy syndrome is likely to be drug resistant
seizures are drug resistant
or treatment is causing intolerable side effects.
This is important because drug resistance should trigger specialist reassessment, not simply repeated medication changes indefinitely.
What can a tertiary epilepsy centre do?
Depending on the person's needs, a tertiary centre may provide access to:
specialist epileptologists
epilepsy specialist nurses
advanced EEG
prolonged video telemetry
specialist epilepsy MRI review
neuropsychology
neuropsychiatry
PET
SPECT
MEG
genetic testing
epilepsy surgery assessment
intracranial EEG or SEEG
neuromodulation
and specialist dietary treatment.
Not everyone needs all these investigations.
They are selected according to the clinical question.
Drug-resistant epilepsy should trigger consideration of surgery
NICE recommends discussing resective epilepsy surgery assessment with people who have drug-resistant epilepsy.
It specifically says people with drug-resistant epilepsy should be referred for consideration of surgical assessment:
including people whose MRI does not show an abnormality. (nice.org.uk)
This is a crucial point.
A normal MRI does not automatically mean:
“Surgery is impossible.”
Why consider surgery so early?
When drug-resistant focal epilepsy comes from a brain region that can be safely treated, surgery may provide a substantially greater possibility of long-term seizure freedom than continuing medication trials alone.
NICE's evidence review concluded that resective epilepsy surgery is the most clinically effective treatment for appropriately selected children, young people and adults with drug-resistant focal epilepsy. (nice.org.uk)
But suitability is highly individual.
The purpose of referral is to find out whether surgery is an option.
Referral for surgery does not mean agreeing to surgery
A surgical assessment is an investigation.
It is not consent to an operation.
The process asks questions such as:
Where do the seizures begin?
Do they consistently begin in one region?
Is that region safe to remove or disconnect?
What important functions are nearby?
How likely is surgery to improve seizure control?
What risks would surgery carry?
The eventual result may be:
surgery is recommended
further investigations are needed
surgery is technically possible but the person chooses not to proceed
surgery would carry too much risk
or another treatment is considered more appropriate.
NICE requires the benefits, risks and assessment process to be discussed with the person and, where appropriate, their family or carers. (nice.org.uk)
A normal MRI does not exclude surgery
This deserves its own section.
Some people have MRI-negative focal epilepsy.
Their seizures may still consistently arise from one brain network.
Specialists can investigate this using combinations of:
video telemetry
repeat high-resolution MRI
specialist MRI review
PET
SPECT
MEG
neuropsychology
and SEEG.
NICE explicitly recommends referral for surgical assessment even when no MRI abnormality has been identified. (nice.org.uk)
Related Information Hub pages:
MRI and Epilepsy: What Doctors Are Looking For
PET, SPECT and MEG in Epilepsy
Video Telemetry: Recording Seizures in Hospital
Some MRI findings justify early referral
NICE also says some people may benefit from tertiary referral before waiting for epilepsy to become drug resistant.
This particularly applies when MRI shows abnormalities associated with a high risk of medication resistance.
NICE examples include:
hippocampal sclerosis
malformations of cortical development
epilepsy-associated low-grade tumours
hypothalamic hamartomas
neuronal migration disorders
tuberous sclerosis complex
vascular malformations
and cerebral contusions from previous head injury.
Early specialist review can prevent potentially useful surgical assessment being delayed for years.
Hippocampal sclerosis
Hippocampal sclerosis is a structural abnormality involving one or both hippocampi and is strongly associated with some forms of temporal-lobe epilepsy.
When seizures continue and the evidence points consistently towards one affected temporal region, specialist surgical assessment may be appropriate.
But the MRI finding alone does not decide whether surgery is possible.
Doctors still need information about:
actual seizure onset
language
memory
the opposite hippocampus
and other important brain functions.
Focal cortical dysplasia
Focal cortical dysplasia — FCD — is an abnormality in cortical development and is an important cause of drug-resistant focal epilepsy.
Some forms are easily visible on MRI.
Others can be extremely subtle or apparently MRI-negative.
Advanced imaging, specialist neuroradiology and intracranial EEG may sometimes be needed to identify the relevant seizure network.
Learning disability should not automatically exclude someone
NICE specifically says that people should not be excluded from consideration of epilepsy surgery solely because of a learning disability or genetic abnormality.
Historically, some people have not been referred because assumptions were made that they would not benefit or would not cope with assessment.
NICE explicitly rejected that approach. (nice.org.uk)
Suitability should be assessed individually.
Genetic epilepsy does not automatically rule out surgery either
A genetic cause can sometimes create a widespread epilepsy network, in which case resective surgery may not be appropriate.
But genetics and focal structural epilepsy can overlap.
A genetic diagnosis should therefore not automatically stop specialist evaluation where the seizure pattern suggests a potentially treatable focal network.
Again, the decision depends on the individual epilepsy rather than one label.
What happens during presurgical assessment?
Testing may include:
detailed seizure history
prolonged video EEG
epilepsy-protocol MRI
specialist neuroradiology review
neuropsychological testing
functional MRI
PET
SPECT
MEG
and sometimes intracranial EEG such as SEEG.
The aim is to build a concordant picture of:
where seizures originate
how they spread
and which important brain functions lie nearby.
Related Information Hub page:
Neuropsychological Testing and Epilepsy
What if resective surgery is not possible?
That still does not mean treatment ends.
Possible reasons surgery may not be suitable include:
seizures begin from several regions
seizure onset cannot be localised sufficiently
the epileptogenic tissue overlaps essential brain function
risks outweigh expected benefits
or the person chooses not to have surgery.
Other approaches may then be considered.
Vagus nerve stimulation
Vagus nerve stimulation — VNS uses an implanted generator connected to the vagus nerve in the neck.
It delivers repeated electrical stimulation intended to reduce seizure frequency or severity.
NICE guidance allows VNS as an add-on treatment when epilepsy remains drug resistant and resective surgery is unsuitable. Epilepsy Action describes it as a treatment aimed primarily at reducing seizures rather than guaranteeing seizure freedom. (epilepsy.org.uk)
VNS does not remove the seizure-generating brain region.
It modifies seizure networks through neuromodulation.
Other neuromodulation treatments
Specialist epilepsy centres may also consider other forms of stimulation for selected people.
Availability depends on:
epilepsy type
individual anatomy
previous treatment
current evidence
NHS commissioning
and the specialist centre.
These treatments continue to develop.
The existence of neuromodulation is another reason the phrase:
“two medicines failed, so nothing else can be done”
is incorrect.
Ketogenic dietary therapy
Ketogenic diets alter the body's main fuel metabolism by producing high levels of ketones.
They are well established for certain epilepsy syndromes and metabolic disorders.
NICE does not recommend ketogenic therapy routinely for every person with drug-resistant epilepsy but allows specialist consideration in selected cases and recommends it specifically within some epilepsy-syndrome pathways. (nice.org.uk)
A medical ketogenic diet is not simply a general low-carbohydrate diet.
It requires specialist supervision because it can affect:
nutrition
growth
blood chemistry
kidney stones
gastrointestinal health
lipids
and medication formulations.
More medication can still be worthwhile
Being labelled drug resistant does not mean doctors stop using antiseizure medicines.
Further medicine trials may still:
stop seizures in some people
reduce frequency
reduce severity
prevent seizure spread
reduce clusters
or improve quality of life.
Medication is also commonly continued alongside:
surgery
VNS
dietary therapy
and other specialist treatments.
The drug-resistant label changes the strategy.
It does not mean medicine has no further role.
Combination therapy
Some people require more than one antiseizure medicine.
This is called polytherapy.
The aim is to combine treatments with complementary mechanisms while avoiding unacceptable adverse effects.
But adding more medicines can increase problems such as:
drowsiness
dizziness
memory problems
balance difficulties
mood effects
and drug interactions.
Specialists therefore need to balance seizure reduction against the burden of treatment.
More tablets are not always better
A person may eventually be taking several antiseizure medicines while continuing to experience seizures.
At that point, repeatedly adding medication without reassessing the broader treatment plan may not be the best approach.
A specialist review can ask:
Is every medicine still helping?
Are doses appropriate?
Are interactions occurring?
Is the diagnosis correct?
Could surgery help?
Could neuromodulation help?
Could a syndrome-specific treatment be more appropriate?
Drug-resistant epilepsy care is about optimising the whole strategy rather than simply maximising the number of medicines.
Could a medicine be making some seizure types worse?
Yes.
Some antiseizure medicines can aggravate particular seizure types or epilepsy syndromes.
For example, NICE warns that certain medicines can exacerbate seizures in specific generalised childhood and adult epilepsy patterns.
This is another reason accurate seizure classification matters when treatment appears unsuccessful. (nice.org.uk)
Treatment should also address the cause where possible
Sometimes an underlying cause offers an additional treatment target.
Examples include:
immune treatment for active autoimmune encephalitis
ketogenic therapy in GLUT1 deficiency
surgery for a suitable structural lesion
treatment of a tumour
or other cause-specific therapies.
Drug resistance should therefore prompt the question:
“Do we fully understand why this person has epilepsy?”
as well as:
“Which medicine should we try next?”
Drug resistance can develop early
Some epilepsy syndromes and structural lesions are associated with a high likelihood of treatment resistance from relatively early in the disease.
This is why modern guidance increasingly discourages waiting through years of unsuccessful treatment before referring someone to a specialist centre.
NICE specifically recommends early referral when MRI findings suggest a high risk of drug resistance. (nice.org.uk)
Drug-resistant epilepsy can change over time
Drug resistance does not necessarily mean seizures can never later become controlled.
Some people experience periods of:
improved control
remission
recurrence
or changing seizure patterns.
A later medicine may occasionally work even after several earlier failures.
Surgery or another treatment may also produce substantial improvement.
The label describes the person's treatment history and current clinical situation.
It is not a prediction that seizure freedom is impossible.
Why seizure counting matters
Determining whether treatment has genuinely failed requires reasonably accurate information about seizures.
But people may not recognise every seizure.
This is especially true for:
sleep seizures
focal impaired-consciousness seizures
absence seizures
brief focal events
and seizures followed by amnesia.
Video telemetry can sometimes reveal that seizure frequency is higher than the diary suggested.
Related Information Hub page:
Video Telemetry: Recording Seizures in Hospital
Why nocturnal seizures still count
A seizure does not stop counting because:
the person slept through it
nobody noticed it
it caused little movement
or it was found only through prolonged monitoring.
If it is a genuine epileptic seizure, it remains relevant to seizure control.
This can alter whether treatment can truly be described as producing seizure freedom.
Drug resistance and quality of life
Seizure count is important, but it is not the only outcome that matters.
Drug-resistant epilepsy can affect:
injuries
sleep
memory
mood
anxiety
relationships
education
employment
independence
and social participation.
Medication side effects may add another burden.
A good specialist review therefore asks not only:
“How many seizures?”
but also:
“What is epilepsy and its treatment doing to this person's life?”
Mental health matters
Living with persistent unpredictable seizures can contribute to:
anxiety
depression
isolation
fear of seizures
reduced confidence
and loss of independence.
Some antiseizure medicines can also influence mood.
Mental-health symptoms should not simply be dismissed as an inevitable consequence of uncontrolled epilepsy.
Epilepsy Action specifically includes mental-health support within its current information on living with drug-resistant epilepsy. (epilepsy.org.uk)
Safety planning matters too
Even while doctors continue working towards better seizure control, people may need practical plans covering:
seizure first aid
prolonged seizures
rescue medication
bathing and water safety
heights
cooking
sleep
medication adherence
and activities where sudden loss of consciousness could cause serious harm.
Safety advice should match the person's actual seizure type rather than imposing the same restrictions on everyone with epilepsy.
Drug resistance does not mean the person caused the treatment failure
This point deserves saying clearly.
Persistent seizures are not evidence that someone:
did not try hard enough
failed treatment
could have controlled the seizures through willpower
or simply failed to avoid enough triggers.
There may be practical issues with medication that need correcting.
But genuine drug-resistant epilepsy reflects the biology of the seizure disorder and its response to treatment.
The medicine failed to provide sustained seizure freedom.
That is different from saying the person failed.
Questions to ask if seizures continue after two medicines
Useful questions include:
Do I now meet the definition of drug-resistant epilepsy?
Were both medicines appropriate for my seizure type?
Did I have adequate treatment trials?
Do we need to reconsider my seizure classification?
Could some of my events be a different type of seizure or a different condition?
Should my MRI be reviewed again by a specialist neuroradiologist?
Should I be referred to a tertiary epilepsy centre?
Should I have video telemetry?
Am I a candidate for epilepsy-surgery assessment?
Does a normal MRI affect whether I can be referred?
Are there other medication combinations worth trying?
Could VNS or another neuromodulation treatment help?
Is dietary therapy relevant to my epilepsy?
Do we know the underlying cause of my epilepsy?
If surgery has never been mentioned
Someone with drug-resistant epilepsy can reasonably ask:
“Should I be referred for an epilepsy-surgery assessment?”
NICE recommends referral for consideration of resective epilepsy-surgery assessment in people with drug-resistant epilepsy even when MRI does not show an abnormality. (nice.org.uk)
Referral does not mean surgery will ultimately be suitable.
It means specialists should assess whether it might be.
The most important message
Drug-resistant epilepsy does not mean:
“There is nothing left to try.”
It has a specific clinical meaning.
The internationally accepted definition is failure of two tolerated, appropriately chosen and appropriately used antiseizure medication schedules to achieve sustained seizure freedom.
Before applying that label, doctors should be confident that:
the diagnosis is correct
seizure types are correctly classified
the medicines were appropriate
treatment trials were adequate
and another explanation for apparent treatment failure has not been missed.
Once drug resistance is recognised, the strategy should broaden.
NICE recommends tertiary specialist referral and consideration of epilepsy-surgery assessment, including for people with a normal MRI.
Other options may include:
further medicine combinations
surgery
VNS or other neuromodulation
specialist dietary treatment
and treatment directed at an underlying cause.
So the significance of the two-medicine threshold is not:
“Treatment has reached the end.”
It is:
“It is time to look beyond routine medication management and make sure every appropriate treatment option has been considered.”
Sources and further reading
International League Against Epilepsy — Drug-Resistant Epilepsy.
The ILAE continues to use its consensus definition based on failure of two tolerated, appropriately selected and appropriately used antiseizure medication schedules to achieve sustained seizure freedom. (ilae.org)
Kwan P and colleagues — Definition of drug-resistant epilepsy: Consensus proposal of the ILAE Commission on Therapeutic Strategies.
The original ILAE framework defines drug resistance, treatment failure and sustained seizure freedom, including the 12-month or three-times-longest-interseizure-interval criterion. (ilae.org)
ILAE — Drug-resistant epilepsy with Professor Patrick Kwan, 2025.
The ILAE's current educational discussion confirms continued use of the two appropriately chosen, adequately used antiseizure medication definition. (ilae.org)
NICE — Epilepsies in children, young people and adults (NG217).
Current NICE guidance recommends referral of people with drug-resistant epilepsy for consideration of resective epilepsy-surgery assessment, including people without MRI abnormalities, and allows early referral when particular structural lesions predict a high risk of treatment resistance. Last updated January 2025. (nice.org.uk)
NICE Quality Standard QS211 — Referral to tertiary specialist services.
NICE states that people with drug-resistant seizures, syndromes likely to be drug resistant, diagnostic uncertainty or intolerable treatment adverse effects should be referred to tertiary epilepsy services, generally to be seen within four weeks. (nice.org.uk)
Epilepsy Action — Drug-resistant epilepsy.
Current UK patient information covering the two-medicine definition, possible reasons for apparent treatment failure, tertiary referral and treatments including epilepsy surgery. (epilepsy.org.uk)
Information reviewed: September 2026.
This page provides general educational information. Ongoing seizures despite appropriate treatment should be reviewed by an epilepsy specialist, and antiseizure medicines should not be stopped or changed without medical advice.
