What SUDEP Is and Is Not

By Dovy Paukstys, Founder, Komori Care

This article is educational and is not medical advice. It does not diagnose, treat, prevent, or cure any condition. Komori is not a medical device, is not FDA-cleared, does not detect seizures, and is not intended for people with epilepsy. Talk with a neurologist or epileptologist about personal medical decisions. Households living with epilepsy should use clinically validated seizure-detection devices as directed by their care team.

SUDEP is a word clinics drop and then walk away from. Families google it at 1am. That's a bad way to learn a mortality term.

I don't have epilepsy. I'm a tech founder in Virginia who reads the source pages because the internet is loud and usually wrong. I opened the 2017 AAN/AES guideline, the AAN patient sheet, NINDS, and CDC. Here's the translation.

What it is

The AAN patient summary (they wrote it with AES) calls SUDEP an uncommon but known complication of epilepsy. Sudden, unexpected death of a person with epilepsy who is otherwise healthy. Cause of death unknown.

NINDS says the same thing with one extra fence: the death is not from injury, drowning, or another known cause. Most cases, not all, happen during or right after a seizure. Their current research pitch is abnormal brain activity hitting heart and breathing. CDC is shorter. Experts think breathing or heart rhythm.

The full guideline made studies use a picky definition. Person had epilepsy. Drowning, trauma, and status epilepticus were out. Death could follow a witnessed seizure. Other competing causes were out.

So it's a category. Not a synonym for "someone with seizures died."

What it is not

Not drowning. Not a fall. Not status epilepticus.

CDC and NINDS treat status as its own emergency. Seizures stacked inside five minutes, or one seizure that will not stop. NINDS says five minutes is enough to damage neurons and those seizures usually don't quit on their own. Call that what it is. Don't file it under SUDEP.

Also not a blog diagnosis. Medical examiners and clinicians apply the definition after they rule other things out.

Also not inevitable. The rates below are population averages. They are not your cousin's number.

And it is not something a consumer monitor prevents. I will say that again later because the internet keeps trying to sell the opposite.

The numbers the guideline actually published

They reviewed 12 Class I incidence studies. Kids 0 to 17: 0.22 SUDEP deaths per 1,000 patient-years. High confidence. Adults: 1.2 per 1,000 patient-years. Low confidence, wide error bars. The adult number is squishier than people admit.

The patient sheet turns that into language a clinician can say out loud.

Kids: rare. 1 in 4,500 children with epilepsy per year. So 4,499 of 4,500 will not be affected that year.

Adults: uncommon. 1 in 1,000 per year. 999 of 1,000 will not.

They wrote it that way on purpose. People hear a catastrophe and inflate it. The panel told doctors to give both sides of the fraction. Honestly, that's the only sane way to say a rare risk.

Averages. Not a household score.

The risk factor they treat as real

Major risk factor, high confidence: generalized tonic-clonic seizures. GTCS. Whole body, convulsions, loss of consciousness. More of those, higher SUDEP risk.

One graded example in the full paper: 3 or more GTCS a month lined up with a 15-fold increase. That translated, in the studies they cited, to as high as 18 deaths per 1,000 patient-years in that frequent-GTCS group. Published figure. Not a forecast for one person.

Seizure freedom, especially freedom from GTCS, was strongly associated with lower SUDEP risk. That's the finding they turned into a counseling line.

A pile of other stuff (nocturnal seizures, drug counts, surgery, vagus nerve stimulation) sat in the low / very low / conflicting bucket. The panel did not crown a "safer drug." I'm not going to either.

What they told clinicians to say

This is a counseling paper. It tells doctors what to discuss. It does not tell families to buy a product. Bah.

Tell parents of kids: rare risk, 1 in 4,500 (Level B).

Tell adults: small risk, 1 in 1,000 (Level B).

If GTCS keep happening, keep managing therapy to cut seizures and SUDEP risk, and weigh the downsides of the next move with the patient (Level B).

Tell people that seizure freedom, particularly from GTCS, is strongly associated with lower SUDEP risk, and that this is more likely with taking the medicine (Level B).

That's the public-health point I will actually stand on. CDC: good seizure control can cut the risk of injury or death. NINDS: people may be able to lower SUDEP risk by taking antiseizure medicine as prescribed and getting the best possible epilepsy care. NINDS also says skipping the prescribed amount, or not being on the right medicine regularly, may raise SUDEP risk, especially if someone is on more than one drug.

None of that is a promise. It's a referral. Neurologist. Epileptologist. Not a gadget.

The Epilepsy Foundation's risk-factor list

The AAN/AES guideline above is the counseling paper. The Epilepsy Foundation publishes the household-facing version, and it is worth reading in their language rather than mine.

Their greatest named risk factor is tonic-clonic seizures. They still print "grand mal" in parentheses so older charts match.

People with nighttime seizures may also be at higher risk.

Missing medications, or not taking seizure medicines as prescribed, can lead to more seizures. That may raise SUDEP risk.

People with poorly controlled epilepsy sit at greater risk than people whose seizures are controlled. The Foundation's public line is that the best way to lower SUDEP risk is to have as few seizures as possible.

They also say people who have only absence or myoclonic seizures are not known to have increased risk for sudden death. That cut belongs to a clinician. Do not self-sort from a blog.

On the same page they list ordinary ways to lower risk by aiming for fewer seizures. Take medicine regularly and at the right dose. See the care team if seizures are not controlled. Ask about surgery, devices, or dietary therapy when medicine is not enough. Know seizure first aid. Be careful around water. Those are their steps. They are clinic and household work. They are not a product category.

Their FAQ on children is equally careful. Whether a child is at risk depends on how severe the epilepsy is and the type of seizures. Some studies found lower rates in children. Others found rates similar to adults. A pediatric neurologist applies that, not a table on the internet.

Why "warning signs of SUDEP" is the wrong search

People search that phrase. I understand why. It is still the wrong question, and the Foundation's own description is what makes it wrong.

They say the person is often found dead in bed, often lying face down, and that SUDEP occurs most often at night or during sleep, when the death is not witnessed. A seizure is not always documented. That is the whole problem with "warning signs": the event the household is afraid of is, by their description, often unwitnessed.

A risk factor is a pattern in a population. It is not a siren in your hallway. None of the list above is a checklist you can use at 2 a.m. to decide "this is the night."

Do not look for a prodrome. Do not treat restlessness, a cough, or a roll as a sign that SUDEP is arriving. Those are ordinary night sounds. A neurologist can talk about this person's seizures. A blog cannot hand you a watch-list.

The Foundation tells people to ask the doctor about SUDEP if the clinic has not brought it up. That conversation is the opposite of a homemade warning-sign chart. It is specific to seizure type, control, and living situation. It is also the only place a personal plan should come from.

Night, face-down, and what the sleep-safety page actually says

SUDEP occurs most often at night or during sleep. The Foundation says the death is often unwitnessed and the person is often found in bed.

That is not a second disease called "night SUDEP." It is the same epilepsy after the lights go out, when there is no one to turn a person, time an event, or call for help. The structural version of that gap is The Night Is Where Epilepsy Care Falls Apart.

Their sleep-safety page is the practical twin. Avoid sleeping on the stomach. Share a bedroom if that is possible. Have someone nearby who can help. Limit extra pillows. Clear sharp objects. If seizures tend to occur only during sleep, ask whether the timing of medicine should change.

"Possible," again, is doing a lot of work. Living alone is a risk conversation, not a moral failing. A neurologist can talk through roommates, partners, and what "nearby" can mean in a real apartment. A blog cannot invent a second adult.

On position: the Foundation says people who die of SUDEP are often found lying face down, and the sleep-safety page says studies seem to suggest that being face down in the bedding may be a factor. That is a position. It is not a rash. It is not a prodrome. You cannot watch for it the way you watch for a fever. Side-lying is the other half of that same sentence — at night it is the habit their sleep-safety page already teaches. Mouth toward the ground. Airway clear. Not on the stomach.

The sentence I will not write

I will not write that a nighttime camera, a wearable, or a "wellness" monitor prevents SUDEP. The 2017 guideline does not hand clinicians a high-level device-prevention claim. If a specific clinic later talks through nighttime precautions with a specific family, that's their call. This page is not that visit.

Komori is not in this story. Not a medical device. Does not detect seizures. Not for people with epilepsy.

Households that need seizure detection should use whatever clinically validated tool their care team actually prescribed.

What to do with this

Take the word to clinic. Ask how this person's seizure types, especially any GTCS history, sit against the AAN/AES counseling. Ask if the current regimen is still the right one. Ask what seizure freedom would even mean here.

If seizures are still happening, CDC's treatment page is blunt. See a neurologist or an epileptologist. That's the move.

lol at the idea that a blog is a plan. It isn't.

FAQ

What does SUDEP stand for?

Sudden unexpected death in epilepsy. AAN: sudden, unexpected death of a person with epilepsy who is otherwise healthy, cause unknown. NINDS: not injury, not drowning, not another known cause.

How common is it?

AAN/AES 2017: typically 1 in 4,500 children with epilepsy per year, 1 in 1,000 adults per year. Most people with epilepsy are not affected in a given year. Your number is a clinic conversation.

Does better seizure control lower the risk?

The guideline tells clinicians that seizure freedom, especially from GTCS, is strongly associated with lower SUDEP risk. CDC and NINDS both tie lower injury and death risk to seizure control and taking prescribed medicine. Associated. Not guaranteed.

Can a home monitor prevent SUDEP?

No. Not on this page. The strongest official counseling point is seizure control under a clinician. A consumer device is not a SUDEP strategy.

Sources

• AAN / AES practice guideline summary (PMC, 2017): https://pmc.ncbi.nlm.nih.gov/articles/PMC5486432/
• AAN guideline product page (reaffirmed Oct 21, 2023): https://www.aan.com/Guidelines/home/GuidelineDetail/850
• AAN patient and family summary: https://www.aan.com/Guidelines/home/GetGuidelineContent/851
• NINDS, Epilepsy and Seizures: https://www.ninds.nih.gov/health-information/disorders/epilepsy-and-seizures
• CDC, Health and Safety Concerns: https://www.cdc.gov/epilepsy/health-safety-concerns/index.html
• CDC, Treatment of Epilepsy: https://www.cdc.gov/epilepsy/treatment/index.html