Epilepsy Syndromes (EEG Correlates)

An epilepsy syndrome is more than a seizure type. It is a recognizable cluster of seizure semiology, EEG signature, typical age of onset, natural history, comorbidities, and increasingly a defined genetic or structural etiology. The 2017 ILAE operational classification and the 2022 ILAE position papers on syndrome definitions elevate the syndrome to a central diagnostic unit because naming the syndrome predicts the EEG one should find, the drugs that will help or harm, the comorbidities to anticipate, and the prognosis. The 2022 framework organizes syndromes deliberately by age of onset, recognizing that the same brain expresses epilepsy differently across the lifespan, and it formalizes the category of developmental and epileptic encephalopathies in which the epileptic activity itself drives cognitive and behavioral impairment beyond what the underlying etiology alone would cause. This module surveys the major syndromes through the lens of their electrographic correlates, beginning with the two great focal epilepsies, moving to the genetic generalized epilepsies, and closing with the age-dependent pediatric syndromes whose EEG patterns are among the most distinctive in all of neurophysiology.

Temporal and frontal lobe epilepsies

Temporal lobe epilepsy (TLE) is the most common focal epilepsy of adults, and mesial TLE with hippocampal sclerosis is its archetype. The interictal EEG shows anterior temporal spikes or sharp waves, maximal at F7/F8, T1/T2, or the basal temporal contacts, often with a stereotyped oblique field that points to mesial structures even when recorded from inferolateral scalp, and the recognized accompanying pattern of temporal intermittent rhythmic delta activity carries nearly the same localizing weight. Ictally, mesial temporal seizures favor a rhythmic 5 to 7 Hz theta discharge that builds over the temporal region, frequently with an electrodecremental or attenuating onset, before propagating to the contralateral temporal lobe. The semiology of an epigastric rising aura, behavioral arrest, oroalimentary automatisms, ipsilateral hand automatisms with contralateral dystonic posturing, and post-ictal dysphasia from the dominant hemisphere maps cleanly onto this propagation. TLE is the paradigm of a surgically remediable epilepsy, and a randomized trial established that anterior temporal resection is superior to continued medical therapy in appropriately selected patients with drug-resistant mesial TLE, which is what makes the precision of its EEG and imaging localization clinically momentous.

Frontal lobe epilepsy (FLE) is the great mimic and the harder localization problem. Its seizures are typically brief, frequent, nocturnal, and explosive, with hyperkinetic or bizarre motor semiology, prominent asymmetric tonic posturing, preserved awareness in some, and rapid evolution to bilateral tonic-clonic activity. The EEG is notoriously unhelpful: the frontal lobe is vast, much of it buried in sulci and on the mesial interhemispheric wall where dipoles are tangential or distant from scalp electrodes, so interictal discharges may be absent, midline, bilateral, or falsely lateralizing, and ictal recordings are frequently obscured by movement and muscle artifact from the violent semiology. Rapid bifrontal and transcallosal propagation can produce an apparent generalized onset that masks the focal driver. The familial nocturnal form, historically called autosomal dominant nocturnal frontal lobe epilepsy and now recognized within the sleep-related hypermotor epilepsies, is associated with variants in neuronal nicotinic acetylcholine receptor subunit genes, most often CHRNA4 and CHRNB2 encoding the alpha-4 and beta-2 subunits and less commonly CHRNA2, with incomplete penetrance on the order of 60 to 80 percent; it was one of the first focal epilepsies shown to have a defined single-gene cause, and it illustrates that a genetic etiology does not imply a generalized epilepsy, since this is a genetically determined focal epilepsy. The condition is frequently misdiagnosed as a parasomnia because the events are nocturnal and bizarre, and the stereotypy of the attacks across nights, their clustering, and the family history are often more diagnostic than the EEG, which is commonly normal or obscured. Recognizing FLE often depends more on the stereotyped clinical pattern, sleep-related clustering, and advanced source localization than on a clean scalp ictal recording.

Localizing focal epilepsies uses the same dipole and field logic taught throughout this course, and the interactive model below lets you rehearse it on a movable source. The lesson it makes concrete is why temporal sources, often with a favorable oblique-to-radial component projecting to the inferolateral scalp, localize more readily than mesial frontal sources, whose tangential midline dipoles displace and obscure the scalp maximum. The same physics that makes mesial TLE a clean localization problem makes mesial FLE a treacherous one.

Idiopathic (genetic) generalized epilepsies

The idiopathic generalized epilepsies (IGE), now often called genetic generalized epilepsies, comprise four overlapping syndromes recognized in the 2022 framework: childhood absence epilepsy, juvenile absence epilepsy, juvenile myoclonic epilepsy (JME), and epilepsy with generalized tonic-clonic seizures alone. Their unifying EEG signature is generalized, bilaterally synchronous, frontally predominant spike-and-wave and polyspike-and-wave on a normal background, generated by the thalamocortical network rather than a focal lesion, which is why the background organization is preserved and the MRI is typically normal. The discharges are exquisitely sensitive to activation: hyperventilation reliably provokes absence discharges, photic stimulation elicits a generalized photoparoxysmal response in a substantial fraction, and sleep deprivation and the early-morning state accentuate the polyspikes of JME, so an activation protocol and a sleep-deprived recording are essential to the diagnosis. The genetics are predominantly complex and polygenic rather than single-gene, with a clear heritable predisposition, contributions from common and rare variants in ion-channel and GABA-receptor genes, and only a minority attributable to a single identifiable mutation; this is an important contrast with the monogenic developmental encephalopathies discussed below. Recognizing IGE is therapeutically decisive because broad-spectrum agents are effective whereas sodium-channel-blocking drugs can paradoxically worsen absence and myoclonic seizures.

The activation procedures themselves deserve emphasis because they are often what converts a normal-appearing routine study into a diagnostic one. Photic stimulation with a strobe across a range of flash frequencies can elicit a photoparoxysmal response, a generalized spike-and-wave or polyspike-and-wave discharge time-locked to the flashes; when this response outlasts the stimulus and is generalized, it is a robust marker of a genetic generalized trait and underlies the clinically familiar photosensitive and reflex epilepsies in which seizures are provoked by flickering light, patterns, or screens. Hyperventilation for three to five minutes reliably provokes the 3 Hz discharges of absence and is so dependable that a failure to activate during good-effort hyperventilation argues against childhood absence epilepsy. Sleep and sleep deprivation add yield across syndromes by lowering the seizure threshold and by unmasking sleep-activated discharges, which is why a complete diagnostic protocol attempts to capture wakefulness, drowsiness, sleep, hyperventilation, and photic stimulation rather than a single awake epoch. The reader who reports a normal study without noting which activations were performed has left the clinician unable to judge how thoroughly the trait was sought.

Childhood absence epilepsy presents the cleanest correlate: frequent brief staring spells with the regular 3 Hz generalized spike-and-wave that begins and ends abruptly and is almost always reproducible with hyperventilation, on a normal background, in an otherwise healthy school-aged child. It frequently remits by adolescence. Juvenile myoclonic epilepsy shows faster, often irregular polyspike-and-wave in the 4 to 6 Hz range, commonly with a photoparoxysmal response and the defining early-morning myoclonic jerks, and it is typically a lifelong condition requiring continued treatment, in contrast to the frequently remitting childhood absence form. Juvenile absence epilepsy sits between them with somewhat less frequent absences and a higher rate of accompanying generalized tonic-clonic seizures. The interictal background in all IGE is normal, an important discriminator from the symptomatic and developmental generalized epilepsies whose backgrounds are abnormal, and this single observation, made before any discharge is even classified, does much of the diagnostic work.

Pediatric and developmental epileptic encephalopathies

Several pediatric syndromes carry EEG signatures so characteristic that the pattern is nearly diagnostic. Self-limited epilepsy with centrotemporal spikes, formerly benign rolandic epilepsy, shows high-amplitude centrotemporal (rolandic) spikes with a characteristic horizontal tangential dipole that is surface-negative centrotemporally and positive frontally, markedly activated by sleep, in a child with infrequent nocturnal hemifacial or oropharyngeal seizures, a normal background, and normal development; it remits by adolescence and often requires no treatment. Its related self-limited syndromes include the occipital epilepsies of childhood. At the opposite end of severity lie the developmental and epileptic encephalopathies, the 2022 category in which the abundant epileptic activity itself contributes to the cognitive and behavioral impairment beyond the underlying cause, so that suppressing the EEG abnormality is not merely seizure control but a potential rescue of development.

Infantile epileptic spasms syndrome, which subsumes the historical West syndrome, is defined by the triad of epileptic spasms, developmental regression, and hypsarrhythmia, a chaotic, very high-amplitude, disorganized pattern of asynchronous multifocal spikes and slow waves with no normal background, often punctuated by an electrodecremental response at the moment of the spasm. Lennox-Gastaut syndrome is characterized by multiple drug-resistant seizure types, intellectual disability, and the electrographic dyad of slow spike-and-wave at 1.5 to 2.5 Hz in wakefulness and runs of generalized paroxysmal fast activity at roughly 10 to 20 Hz in sleep, the latter correlating with the tonic seizures that are the syndrome's most dangerous feature. Dravet syndrome, a genetic developmental and epileptic encephalopathy classically caused by loss-of-function variants in the sodium-channel gene SCN1A, begins with prolonged fever-sensitive and often hemiclonic seizures in an infant with an initially normal EEG and normal development, which later deteriorates with generalized and multifocal discharges, photosensitivity, and developmental plateau or regression. The SCN1A mechanism is the textbook case where genetics dictates therapy: because the mutation impairs interneuron sodium channels and sodium-channel blockers further suppress those inhibitory neurons, drugs such as carbamazepine, phenytoin, and lamotrigine worsen Dravet seizures, while broad-spectrum agents and syndrome-specific therapies are used instead.

A distinct and instructive pediatric category is the spectrum of developmental and/or epileptic encephalopathy with spike-wave activation in sleep (D/EE-SWAS), the 2022 ILAE term for the entity long known as continuous spike-and-wave during slow sleep (CSWS) or electrical status epilepticus in sleep (ESES). Its defining EEG signature is a striking activation of epileptiform discharges during non-REM sleep, classically generalized spike-and-wave that becomes nearly continuous through a large fraction of slow-wave sleep, in a child whose discharges may be far sparser when awake. The clinical hallmark is regression or stagnation of cognition, language, behavior, or motor skills that tracks the sleep-activated discharges, the model case of epileptic activity itself driving the encephalopathy, and the principle that suppressing the EEG abnormality may rescue development applies here with particular force. Landau-Kleffner syndrome, an acquired epileptic aphasia with auditory verbal agnosia, sits within this spectrum as the variant in which the regression is predominantly linguistic. The lesson for the reader is that a sleep recording is not optional in a child with unexplained developmental regression, because the diagnostic abnormality can be invisible in the waking state and revealed only when sleep unmasks it.

The molecular genetics of these encephalopathies has moved from research to routine practice in the 2020s. Gene-panel and exome or genome sequencing now identify a causative variant in a substantial proportion of developmental and epileptic encephalopathies, naming genes such as SCN1A, SCN2A, SCN8A, KCNQ2, CDKL5, STXBP1, PCDH19, and many others, and the result increasingly carries direct therapeutic consequences. The clearest examples are precision therapies: KCNQ2 and certain other channelopathies may respond to drugs targeting the affected channel, SCN2A and SCN8A gain-of-function variants that present early can respond to sodium-channel blockers while loss-of-function variants do not, and antisense oligonucleotide and gene-targeted therapies for specific genetic epilepsies have entered clinical development. The EEG does not diagnose the gene, but the electroclinical syndrome the EEG helps define is what triggers the targeted genetic testing, and the two together now drive management in a way that was unavailable a decade ago.

Across all these syndromes the reasoning is integrative and Bayesian: the age of onset sets the prior, the EEG signature provides the likelihood, and the seizure semiology and background organization refine the posterior into a syndrome. A 3 Hz generalized spike-and-wave on a normal background in a seven-year-old with staring spells is overwhelmingly childhood absence epilepsy; the same discharge frequency on a slow, disorganized background in a developmentally delayed child shifts the diagnosis toward a symptomatic or developmental generalized epilepsy with a very different prognosis and treatment. Reading the background, the discharge morphology, the activation responses, the state dependence, and the age together, rather than fixating on any single feature, is the expert habit that converts a pattern into a syndrome and a syndrome into the right treatment and the right genetic test. The most common failure modes are anchoring on a single discharge while ignoring an abnormal background, missing the activation procedures that would have revealed a generalized trait, and forcing a clearly age-specific syndrome into an age-inappropriate diagnosis.