Landau-Kleffner Syndrome

Abstract

Landau-Kleffner syndrome is a rare childhood epileptic encephalopathy in which a previously normal child loses the ability to understand spoken language over weeks to months. The core deficit is a verbal auditory agnosia: the ear works, but the brain can no longer decode speech into meaning, and expressive language then withers for want of comprehension. The engine of the loss is not the occasional overt seizure but a storm of epileptiform discharges that floods slow-wave sleep, so that a child whose daytime seizures are mild or absent may still regress severely. It belongs to cognitive psychology because it shows that a specific, high-level function, language comprehension, can be dismantled by abnormal electrical activity in the tissue that supports it, and that the developing brain's timing matters as much as the lesion.

Keywords: Landau-Kleffner syndrome, acquired epileptic aphasia, verbal auditory agnosia, epileptic encephalopathy, spike-wave index

Landau-Kleffner syndrome (LKS), or acquired epileptic aphasia, is a rare disorder of childhood in which a child with previously normal development loses acquired language in close association with an epileptic electroencephalogram (EEG) (Landau & Kleffner, 1957). The regression usually begins between ages three and eight, and its first and most striking feature is a failure of *comprehension*: the child stops responding to speech, and is often first suspected of having gone deaf, before it becomes clear that hearing is intact and it is the decoding of speech that has failed (Stefanatos, 2011). What makes LKS a subject for cognitive psychology, and not paediatric neurology alone, is what it demonstrates: that continuous epileptiform activity, concentrated in the auditory-language cortex and vastly amplified during sleep, can selectively disassemble a specific cognitive faculty in a child whose brain was, until then, developing normally.

What Landau-Kleffner Syndrome Is

Landau-Kleffner syndrome is one of the childhood epileptic encephalopathies: disorders in which the epileptic activity itself, and not merely an underlying brain lesion, is thought to drive a progressive loss of cognitive function (Specchio et al., 2022). It was first described in 1957 in six children who lost previously acquired speech alongside a paroxysmal, epileptic EEG, and the eponym preserves the two clinicians who paired the language loss with the electrical disturbance (Landau & Kleffner, 1957). It is rare, but it has an outsized place in cognitive neuroscience because of what it isolates: an acquired, often reversible loss of a single high-level faculty in an otherwise developing child.

Three features define the syndrome and are worth holding apart. The first is the *aphasia*: an acquired loss of language in a child who had already learned to speak, beginning characteristically with comprehension. The second is the *epileptic EEG*: abnormal discharges, maximal over the temporal regions, that are dramatically activated by sleep. The third is the *dissociation* between the two visible layers of the disorder — the overt seizures, which may be infrequent, mild, or entirely absent, and the language and behavioural regression, which can be severe regardless (Stefanatos, 2011). It is this dissociation that first alerted clinicians that the damage was being done not by the occasional convulsion but by the continuous, largely hidden, electrical activity beneath it.

LKS is now understood as one point on a continuum, the epilepsy-aphasia spectrum, that runs from the common and benign self-limited epilepsy with centrotemporal spikes at one end to LKS and the broader encephalopathy with continuous spike-and-wave during sleep at the other (Bolsterli Heinzle, 2017). Placing LKS on that spectrum matters, because it links a rare, devastating regression to a common, mild childhood epilepsy through a shared mechanism and, in some children, a shared genetic cause. The spectrum is formally recognised within the International League Against Epilepsy's classification of the epilepsies, which groups these syndromes by their shared electroclinical features rather than by seizure type alone (Scheffer et al., 2017). Because the seizures of LKS, when present, are classified by the same operational scheme applied to any focal epilepsy (Fisher et al., 2017), the distinctiveness of the syndrome lies not in the seizure type but in the cognitive regression it produces. Table 1 sets LKS against its neighbours on that continuum.

Table 1

Landau-Kleffner Syndrome on the Epilepsy-Aphasia Spectrum

How the Language Loss Happens

The regression of LKS is driven by sleep. In wakefulness the EEG may show only scattered temporal spikes, but as the child enters slow-wave (non-REM) sleep the discharges multiply until they occupy a large fraction of the record, a pattern shared with the encephalopathy known as continuous spike-and-wave during slow sleep, or CSWS, and its electrographic hallmark, electrical status epilepticus in sleep (ESES) (Rubboli & Tassinari, 2019). The prevailing account is that this near-continuous nocturnal discharge, concentrated in the perisylvian auditory-language cortex, disrupts the normal work that sleep does in consolidating the day's learning and maintaining synaptic function, so that language is degraded night after night.

The clinical course follows from this mechanism. Because the deficit begins in the auditory decoding of speech, the earliest sign is usually a loss of comprehension that can be mistaken for deafness; expressive language then deteriorates, since a child who can no longer understand speech loses the feedback that sustains their own (Deonna & Roulet-Perez, 2010). The tempo varies: some children decline abruptly over days, others in a fluctuating, stepwise fashion over months, with partial recoveries that track the waxing and waning of the EEG. That the language can return when the discharges subside — and relapse when they return — is the strongest clinical evidence that the epileptic activity is the proximate cause of the aphasia, not an incidental accompaniment to some fixed lesion (Genton & Guerrini, 1993).

The Cognitive Cost of an Epileptic Aphasia

The signature deficit of LKS is a *verbal auditory agnosia*: a specific inability to recognise the meaning of speech sounds despite intact hearing, intact non-verbal sound recognition, and, at least initially, intact general intelligence (Stefanatos, 2011). The child may still hear and even localise environmental sounds — a doorbell, a dog — while failing entirely to decode spoken words, a dissociation that places the lesion not in the ear or the brainstem but in the cortical machinery that turns an acoustic signal into linguistic meaning. It is, in effect, an acquired disorder of auditory perception at its highest, most language-specific level, closely related to the adult syndrome of pure word deafness but arising in a developing brain (Agnosia).

The cost rarely stays confined to language. Because the same sleep-activated discharges disrupt the restorative and consolidating functions of sleep, many children develop a broader picture: attentional and behavioural disturbance, hyperactivity, and in some a regression with autistic features, especially when onset is early (Deonna & Roulet-Perez, 2010). Clinicians recognised early that the syndrome is *much more than aphasia and epilepsy*, entailing a wider neuropsychological and behavioural disturbance than the language loss alone would suggest (Eslava-Cobos & Mejia, 1992). This overlap has fed a long debate about the boundary between LKS and acquired autistic regression, and about how far epileptiform activity contributes to developmental regression more generally. The prognosis for the epilepsy is paradoxically good — seizures and the EEG abnormality almost always remit by adolescence — but the prognosis for *language* is not: a substantial proportion of children are left with lasting deficits in comprehension and expression, and the younger the onset and the longer the active phase, the worse the eventual language outcome (Caraballo et al., 2014).

Worked Example

Because the damage in LKS is done by sleep-activated discharges rather than by countable seizures, its severity is quantified not by seizure frequency but by how much of sleep the epileptiform activity consumes. The standard measure is the *spike-wave index* (SWI): the percentage of slow-wave (non-REM) sleep occupied by spike-and-wave discharges (Rubboli & Tassinari, 2019). Suppose an overnight EEG records $400$ minutes of slow-wave sleep, and spike-and-wave discharges are present during $340$ of those minutes. The spike-wave index is the ratio of the two, expressed as a percentage:

$$\text{SWI} = \frac{T_\text{spike-wave}}{T_\text{NREM}} \times 100\% = \frac{340}{400} \times 100\% = 85\%$$

An index of $85\%$ falls in the range historically used to define electrical status epilepticus in sleep, the extreme electrographic state associated with LKS and CSWS (Rubboli & Tassinari, 2019). The exact threshold is debated — different groups have used cut-offs from roughly $50\%$ to $85\%$, and the index is only a summary of a richer picture — but the logic is what matters for cognitive psychology: the clinically relevant quantity is the *burden of discharge during sleep*, not the number of daytime seizures. A child with an SWI of $85\%$ and no visible seizures at all can be regressing faster than a child with occasional convulsions and a quiet sleep EEG. The measure makes concrete the syndrome's central lesson, that the harm is proportional to the hidden electrical activity, and it is why treatment in LKS aims explicitly at suppressing the nocturnal discharges, not merely at stopping overt seizures.

Discussion

The scientific interest of Landau-Kleffner syndrome lies in the clean dissociations it forces upon us. It separates *hearing* from the *comprehension of speech*, showing that the two are supported by distinct machinery that can fail independently. It separates the *overt seizure* from the *epileptic process*, showing that a brain can be doing continuous cognitive damage while producing few or no visible convulsions. And it separates the *epilepsy* from the *aphasia* in time, because the seizures reliably remit while the language deficit may not, so that the electrical disorder can leave a permanent cognitive scar after it has itself departed (Stefanatos, 2011).

The three demonstrations on this page track that logic. The language-regression display shows how comprehension and expression fall and partially recover as the sleep EEG waxes and wanes; the spike-wave-index tool lets the reader compute the burden of nocturnal discharge and see why it, rather than seizure count, indexes severity; and the verbal-auditory-agnosia display shows how an intact ear can deliver sound to a cortex that can no longer turn it into words. What remains genuinely open is the mechanism by which sleep-activated discharge degrades a specific function: whether it acts by disrupting sleep-dependent consolidation, by driving maladaptive plasticity in the perisylvian cortex during a sensitive developmental window, or by some combination, and why the same electrical state produces language loss in one child and a broader regression in another (Deonna & Roulet-Perez, 2010).

Cognitive Implications

The deepest implication of LKS for cognition is that high-level faculties are not merely localised but *developmentally scheduled*, and that abnormal activity during a sensitive period can dismantle a faculty that a fixed lesion in an adult might spare. The same discharge burden that produces a transient aphasia in an adult can, in a young child whose language networks are still being tuned, produce a lasting deficit, because it corrupts the very process by which those networks are built (Caraballo et al., 2014). Language comprehension, in this light, is not a fixed capacity switched on once and for all but a structure that must be maintained and consolidated, night by night, and that can be eroded when the maintenance is disrupted.

LKS also reframes what an epileptic discharge *does*. We tend to picture a seizure as an event that seizes control of behaviour, but the discharges of LKS mostly do no such thing; they run quietly through sleep and are legible only as a corruption of a cognitive function measured the following day. That a hidden electrical process can selectively subtract the comprehension of speech, while leaving hearing, non-verbal cognition, and even the recognition of environmental sound intact, is a striking confirmation that the brain decomposes a seemingly unitary ability, understanding what is said, into separable components that abnormal activity can pick off one at a time (Stefanatos, 2011).

Current Directions

The most consequential recent development is genetic. Mutations in *GRIN2A*, the gene encoding the GluN2A subunit of the NMDA glutamate receptor, were identified as a major cause of the epilepsy-aphasia spectrum, present in a substantial minority of children with LKS and related disorders and providing the syndrome's first firm molecular anchor (Lesca et al., 2013). Because the NMDA receptor is central to synaptic plasticity and to the developmental tuning of cortical circuits, the finding links the clinical picture to a plausible cellular mechanism and has reframed LKS as, in part, a channelopathy of the glutamatergic synapse; the genetics of the spectrum, and the role of *GRIN2A*, remain an active field (Lesca et al., 2019).

Alongside the genetics, two clinical fronts are active. The first is nosological: the 2022 International League Against Epilepsy classification of childhood-onset syndromes has sharpened the definitions of LKS, CSWS, and the surrounding spectrum, which matters because a consistent definition is the precondition for comparing treatments across centres (Specchio et al., 2022). The second is therapeutic. Treatment is aimed squarely at suppressing the sleep-activated discharge — with corticosteroids, certain antiseizure medications, or, in selected drug-resistant cases with a well-localised focus, the surgical technique of multiple subpial transection pioneered for this syndrome (Sinclair & Snyder, 2005), (Morrell et al., 1995) — and the open question is which children benefit, and how early intervention must come to protect language.

Common Misconceptions

The child has gone deaf.

No. Hearing is intact; the deficit is central. The child can still detect and often localise environmental sounds but can no longer decode speech into meaning, a verbal auditory agnosia that is frequently mistaken for deafness at first (Stefanatos, 2011).

The damage is done by the visible seizures.

In LKS overt seizures are often mild or absent. The regression is driven by continuous epileptiform discharge that surges during sleep, so a child can regress severely with few or no visible convulsions (Rubboli & Tassinari, 2019).

Because the epilepsy remits, the child recovers fully.

The seizures and the EEG abnormality almost always resolve by adolescence, but language does not reliably return. A substantial proportion of children are left with lasting comprehension and expression deficits, especially with early onset and a long active phase (Caraballo et al., 2014).

Landau-Kleffner syndrome is just a form of autism.

The two can overlap, and early-onset regression with autistic features has fuelled real debate, but LKS is defined by an acquired, epilepsy-linked loss of previously normal language with a characteristic sleep-activated EEG, not by the developmental profile of autism (Deonna & Roulet-Perez, 2010).

Glossary

Acquired epileptic aphasia.

The descriptive name for Landau-Kleffner syndrome: a loss of previously acquired language occurring in association with an epileptic EEG in a child who had developed normally.

Agnosia.

A failure to recognise the meaning of a sensory input despite intact sensation; in LKS it is specifically verbal auditory agnosia, a failure to recognise speech.

Aphasia.

An acquired impairment of language — its comprehension, production, or both — caused by brain dysfunction rather than by a deficit of hearing, articulation, or intellect.

Continuous spike-and-wave during slow sleep (CSWS).

An epileptic encephalopathy in which epileptiform discharges dominate slow-wave sleep, producing cognitive and behavioural regression; LKS is regarded as closely related to, or part of, this condition.

Electrical status epilepticus in sleep (ESES).

The electrographic hallmark of CSWS: near-continuous spike-and-wave activity that appears when the child enters slow-wave sleep, quantified by the spike-wave index.

Electroencephalogram (EEG).

A recording of the brain's electrical activity from scalp electrodes; in LKS it shows temporally predominant epileptiform discharges that are dramatically activated by sleep.

Epilepsy-aphasia spectrum.

A continuum of related childhood disorders sharing sleep-activated centrotemporal discharges, running from benign self-limited epilepsy with centrotemporal spikes to LKS and CSWS.

Epileptic encephalopathy.

A condition in which the epileptic activity itself, rather than an underlying static lesion, is thought to cause progressive impairment of cognition and behaviour.

GRIN2A.

A gene encoding the GluN2A subunit of the NMDA glutamate receptor; mutations in it are a major identified genetic cause of the epilepsy-aphasia spectrum, including some cases of LKS.

Multiple subpial transection.

A surgical technique that cuts horizontal cortical fibres while sparing vertical columns, devised to interrupt the spread of epileptic discharge in eloquent cortex without removing it; developed for LKS.

NMDA receptor.

A glutamate-gated ion channel central to synaptic plasticity and cortical development; the GRIN2A mutations of the epilepsy-aphasia spectrum alter one of its subunits.

Regression.

The loss of a previously acquired skill; in LKS it is the loss of established language, distinguishing the syndrome from a failure to acquire language in the first place.

Slow-wave sleep.

The deep, non-REM stage of sleep marked by high-amplitude slow EEG waves; in LKS it is the state in which epileptiform discharges surge to their maximum.

Spike-wave index (SWI).

The percentage of slow-wave sleep occupied by spike-and-wave discharges; the standard measure of discharge burden in LKS and CSWS, and the index of severity used in place of seizure count.

Verbal auditory agnosia.

The inability to recognise the meaning of speech sounds despite intact hearing and intact recognition of non-verbal sound; the core deficit of Landau-Kleffner syndrome.

Key Researchers

Frank R. Kleffner

American speech-language pathologist and co-author of the founding 1957 report; the language-regression half of the eponym is his contribution, framing the disorder as one of acquired verbal auditory agnosia rather than a primary defect of speech. Wikidata

William M. Landau

American neurologist who, with Frank Kleffner, gave the first description of the syndrome in 1957: children who lost previously normal language after developing a convulsive disorder and an epileptic EEG. Wikidata

Gaetan Lesca

French clinical geneticist who led the identification of GRIN2A mutations as a major genetic cause of the epilepsy-aphasia spectrum, giving Landau-Kleffner syndrome and its relatives their first firm molecular anchor. ORCID

Ingrid Scheffer

Australian paediatric neurologist and epilepsy geneticist who co-chaired the International League Against Epilepsy task forces behind the modern classification of the epilepsies and of childhood-onset syndromes, the framework that situates LKS on the epilepsy-aphasia spectrum. ORCID

Frequently Asked Questions

What is Landau-Kleffner syndrome?

It is a rare childhood disorder, also called acquired epileptic aphasia, in which a child who had developed normally loses language in association with an epileptic EEG. The loss usually begins between ages three and eight and starts with a failure to understand speech.

Is the child deaf?

No. Hearing is intact. The child can still detect and often locate environmental sounds but can no longer decode spoken words into meaning. This central deficit, a verbal auditory agnosia, is often mistaken for deafness when the regression first begins.

What causes the language loss?

The proximate cause is continuous epileptiform discharge that surges during slow-wave sleep over the auditory-language cortex. This near-continuous nocturnal activity is thought to disrupt the consolidation and maintenance of language, degrading comprehension night after night.

Does the child have obvious seizures?

Often not. Overt seizures in LKS may be infrequent, mild, or entirely absent. The severity of the regression tracks the hidden burden of discharge during sleep, measured by the spike-wave index, rather than the number of visible seizures.

What is the spike-wave index?

It is the percentage of slow-wave sleep occupied by spike-and-wave discharges. It is the standard measure of severity in LKS because the harm is done by sleep-activated activity, so the burden during sleep matters more than any daytime seizure count.

Does Landau-Kleffner syndrome run in families?

Most cases are not clearly inherited, but mutations in the GRIN2A gene, which encodes part of the NMDA glutamate receptor, have been found in a substantial minority of children on the epilepsy-aphasia spectrum, providing the syndrome's first firm genetic cause.

Do children recover?

The epilepsy has a good prognosis: seizures and the EEG abnormality almost always remit by adolescence. Language recovery is less certain. Many children are left with lasting comprehension and expression deficits, and earlier onset with a longer active phase predicts a worse language outcome.

How is it treated?

Treatment aims to suppress the sleep-activated discharge rather than merely stop overt seizures, using corticosteroids or certain antiseizure medications, and, in selected drug-resistant cases with a well-localised focus, the surgical technique of multiple subpial transection developed for this syndrome.

Support Organizations

References