Epilepsy is a medical term used to describe anyone who has seizures in the broadest sense, but clinical practice draws a sharper line. A single seizure caused by a temporary trigger, such as low blood sugar or a high fever, is not epilepsy. The diagnosis generally requires at least two unprovoked seizures more than 24 hours apart, or one seizure plus a high risk of recurrence.

That gap between the plain-language definition and the clinical one causes real confusion. Patients are told they "have epilepsy" after one event, or told they do not, and neither explanation lands. The sections below unpack what a seizure actually is at the cellular level, why the 24-hour rule exists, where the definition breaks down, and what to do with it.

What a seizure actually is at the cellular level

A seizure is a sudden, self-sustaining burst of synchronized electrical activity in a population of neurons. Normal brain activity is noisy but balanced: excitatory signals push neurons toward firing, and inhibitory signals, chiefly from GABA-releasing interneurons, hold them back. A seizure begins when that balance tips locally and the tipping point becomes self-reinforcing.

Three mechanisms drive the tipping point. First, excessive glutamate release or heightened sensitivity of glutamate receptors makes neurons easier to recruit. Second, loss of GABA-mediated inhibition removes the brake. Third, changes in ion channels, particularly sodium and potassium channels that govern how quickly a neuron resets after firing, let a group of cells fire in lockstep rather than at random.

Once enough neurons fire together, the synchronized wave can spread through connected circuits. How it spreads determines the symptoms. A discharge confined to the motor cortex produces jerking on one side of the body. A discharge that spreads to both hemispheres and recruits the thalamus produces the loss of consciousness typical of a generalized tonic-clonic seizure.

Seizures stop through active termination, not exhaustion alone. The brain releases inhibitory signals, alters ion concentrations around neurons, and quiets the recruited network. This is why most seizures last under two minutes. A seizure that does not terminate on its own is a medical emergency.

The plain version: a seizure is a group of brain cells firing together when they should be firing independently, and the brain usually has built-in brakes that eventually stop it.

How often seizures and epilepsy occur, and where the thresholds sit

Roughly 1 in 10 people will have at least one seizure in their lifetime, according to the World Health Organization, but far fewer develop epilepsy. Global estimates put epilepsy prevalence near 50 million people, and the CDC places active epilepsy in the United States at about 1 percent of adults and a somewhat lower share of children.

The incidence curve is bimodal. Seizures cluster in early childhood, when fever and developmental factors dominate, and again after age 60, when stroke, tumors, and neurodegenerative disease raise risk. Between those poles, incidence is comparatively low.

The 24-hour rule is not arbitrary. After a first unprovoked seizure, recurrence risk over the following two years sits in a broad range depending on the cause, EEG findings, and imaging. A normal EEG and normal MRI after a single event implies lower risk; an abnormal EEG with epileptiform discharges implies substantially higher risk. The International League Against Epilepsy formalized the current definition in 2014, allowing diagnosis after one seizure when recurrence risk is high enough, roughly comparable to the risk after two unprovoked seizures.

A key number for context: status epilepticus, a seizure lasting longer than five minutes or repeated seizures without recovery between them, is a neurologic emergency with meaningful mortality. The five-minute threshold is operational, not biologic. It tells bystanders when to call emergency services, not when the brain begins to suffer.

Where the definition breaks down

The plain definition, anyone who has seizures has epilepsy, fails in several specific places.

Provoked seizures. A seizure caused by an identifiable transient trigger, such as severe hypoglycemia, alcohol withdrawal, or an acute head injury, is not epilepsy under the clinical definition because the cause is external and potentially reversible. Remove the trigger, and the seizure risk may return to baseline.

Febrile seizures. Roughly 2 to 5 percent of children under 5 have a seizure with fever. The large majority never develop epilepsy. Treating every febrile seizure as epilepsy would mislabel millions of children.

Acute symptomatic seizures in adults. Seizures in the first week after a stroke or traumatic brain injury are common and are usually managed differently from chronic epilepsy, because the underlying injury may stabilize.

Single unprovoked seizure. Some patients meet criteria for epilepsy after one event because their recurrence risk is high. Others do not. The label depends on evidence, not on the number of events alone.

Psychogenic nonepileptic seizures. These events resemble seizures but are not caused by abnormal electrical activity. They are a real and distressing condition, and they require different treatment. Misdiagnosis in either direction is common.

Misconceptions that shape real decisions

The most damaging misconception is that epilepsy and seizures are interchangeable, so a single event means a lifelong diagnosis. The clinical definition exists precisely to prevent that overreach.

A second misconception is that epilepsy is always inherited. Genetic factors matter in many cases, but acquired causes, including stroke, brain tumors, traumatic brain injury, and infections such as neurocysticercosis, account for a large share of cases worldwide.

A third is that seizures always involve convulsions and lost consciousness. Focal seizures can produce brief sensory distortions, unexplained fear, lip-smacking, or a few seconds of unresponsiveness that observers miss entirely. Some people live with epilepsy for years before diagnosis because their events do not match the public image.

A counterintuitive professional caveat: a normal EEG does not rule out epilepsy. Routine EEGs capture only 20 to 30 minutes of activity, and many people with epilepsy have normal interictal recordings. A negative EEG shifts probability but does not close the question. Conversely, an abnormal EEG in someone without clinical events can reflect a predisposition rather than active epilepsy.

Practical actions and red flags

If you witness a seizure, the priorities are safety and time. Ease the person to the floor, turn them onto their side, clear hard objects away, and time the event. Do not put anything in the mouth. Do not restrain the limbs.

Call emergency services if any of the following apply:

The seizure lasts more than five minutes, or a second seizure begins before the person recovers. The person is pregnant, injured, or does not regain consciousness after the event ends. The seizure occurred in water or involved breathing difficulty. It is the person's first known seizure. The person has known epilepsy but the seizure differs from their usual pattern.

For evaluation after a first event, the standard workup includes a detailed history from witnesses, a neurologic exam, blood glucose and electrolyte testing, and usually an EEG and brain MRI. The history matters more than any single test, which is why bystander accounts are valuable.

Scenario Typical classification Usual next step
Seizure with fever in a child under 5 Febrile seizure Pediatric evaluation; most need no long-term treatment
Seizure during severe hypoglycemia Provoked (acute) seizure Correct glucose; treat underlying cause
Two unprovoked seizures 3 months apart Epilepsy Neurology referral; discuss antiseizure medication
Seizure in first week after stroke Acute symptomatic seizure Treat acutely; reassess chronic risk
One unprovoked seizure, abnormal EEG Possible epilepsy (high risk) Neurology referral; shared decision on treatment
Episodes with normal EEG, atypical features Possible nonepileptic events Video-EEG monitoring

What the workup looks like over time

Time point What happens Why it matters
Moment of event Bystander timing and description Duration and features drive urgency
Emergency evaluation Glucose, electrolytes, neuro exam Rules out reversible triggers
Within days Routine EEG, brain MRI Identifies epileptiform activity and structural causes
1 to 2 weeks Neurology follow-up Confirms or revises diagnosis
1 to 3 months Response to treatment, repeat EEG if needed Guides medication decisions
6 to 12 months Seizure-free interval assessment Informs driving and work restrictions
2 to 5 years Reassessment of ongoing need for medication Some patients can taper under supervision

Limits of the definition and of this article

The two-unprovoked-seizures rule is a useful threshold, not a law of nature. It reflects recurrence risk, and recurrence risk varies widely by cause and by individual. Two people with the same seizure count can have very different prognoses. The definition also depends on accurate history, which is often incomplete.

Diagnosis and treatment decisions require a clinician who can integrate history, examination, imaging, and EEG. Nothing here substitutes for that assessment, and antiseizure medication decisions in particular depend on seizure type, comorbidities, and patient-specific factors.

This article is general educational information, not a diagnosis, treatment plan, or medical advice. It does not promise any outcome. If you or someone else has had a seizure, or if a seizure lasts more than five minutes, seek evaluation from a qualified local clinician or call local emergency services immediately.