l kleffner ingles

Epileptic Encephalopathy of Late Childhood

Landau-Kleffner Syndrome and the Syndrome of Continuous Spikes and WavesDuring Slow-Wave Sleep

Michael C. Smith and Thomas J. Hoeppner

Department of Neurosciences, Rush University Medical Center, Chicago, Illinois, U.S.A.

Landau-Kleffner syndrome (LKS) and the syndrome of continuous spikes and wavesduring slow wave sleep (CSWS) are two points on the spectrum of functionalchildhood epileptic encephalopathies. They are characterized by a severe paroxysmalEEG disturbance that may permanently alter the critical synaptogenesis by strength-ening synaptic contacts that should have been naturally “pruned.” The much morecommon benign epilepsy with centrotemporal spikes is also related to LKS and CSWSby a common pathophysiology. Although prognosis in LKS and CSWS for seizurecontrol is good, cognitive function declines and permanent neuropsychologic dysfunc-tion is seen in many cases. This permanent damage is most evident in those patientswho had early-onset EEG abnormality and a prolonged active phase of continuousspike-and-wave discharges during sleep. If the active phase of paroxysmal activitypersists for over 2 to 3 years, even successful treatment does not resolve neuropsy-chologic sequelae. In LKS, the paroxysmal activity permanently affects the posteriortemporal area and results in auditory agnosia and language deficits; in CSWS, thefrontal lobes are more involved and other cognitive disturbances predominate. Ag-gressive treatment should include high-dose antiepileptic drugs, corticosteroids, andsurgery in specific cases Key Words: Landau-Kleffner syndrome—Continuous spikesand waves during slow-wave sleep—Age-dependent epileptic encephalopathy—Epi-leptic aphasia—Electroecephalography—Epilepsy in children.

The two syndromes, Landau-Kleffner syndrome(LKS) and the syndrome of continuous spikes and wavesduring slow-wave sleep (CSWS), have much in com-mon, although the International League Against Epilepsy(1989) and the World Health Organization (1992) haveidentified these disorders separately. Recent research isleading to a consensus that there may be common fea-tures in the pathophysiology of the two syndromes (De-onna, 1991; Hirsch et al., 1995; Maquet et al., 1995). In1993, a symposium in Venice was held that expoundedthis theory, and the results were published by Beauman-oir et al. in 1995. Since this publication, there is a

growing opinion that LKS could be a subtype of CSWS(Morrell, 1995).

Both CSWS and LKS are age-related epileptic en-cephalopathies, linking them to other well-known epi-leptic encephalopathies such as West or Lennox-Gastautsyndromes. However, they do differ in that both LKSand CSWS have a milder course and they represent afunctional disturbance induced by age-related and usu-ally self-limited paroxysmal activity.

HISTORICAL BACKGROUND

The LKS is named appropriately for William Landauand Frank Kleffner, who together in 1957 reported sixchildren with a syndrome characterized as a convulsivedisorder, along with acquired aphasia (Landau and Klef-

Address correspondence and reprint requests to Dr. Michael C.Smith, Department of Neurosciences, Rush University Medical Center,Chicago, IL, 60612–3833, U.S.A.

Journal of Clinical Neurophysiology20(6):462–472, Lippincott Williams & Wilkins, Inc., Philadelphia© 2003 American Clinical Neurophysiology Society

462

fner, 1957). In five children, the developmental aphasiawas related to the convulsive disorder, but the remainingchild became aphasic and hemiplegic after a head injury.The aphasia in these children developed over days tomonths and persisted from 2 weeks to several years. Theseizures included petit mal, grand mal, and myoclonicseizures. The paroxysmal abnormality in the EEG wasusually bilateral but was most prominent over the tem-poral lobes.

These authors suspected a relation between the sever-ity of the paroxysmal discharges on EEG and the severityof language disturbance. They also noticed that theprognosis was good and the seizures were easily con-trolled. They proposed that the paroxysmal discharge inthe area of the brain that was responsible for languageresulted in language dysfunction.

Patry and colleagues in 1971 described “subclinicalelectrical status epilepticus” induced by sleep in sixchildren, 7 to 12 years of age, who displayed seizures,cognitive decline, and language dysfunction (Patry et al.,1971). All had almost continuous spike-and-wave duringnonrapid eye movement (NREM) sleep. They had atonic,generalized tonic-clonic, convulsive, clonic, and atypicalabsence seizures. Two failed to acquire language, andfive were mentally retarded. The degree of mental retar-dation was directly related to the age of onset. Theauthors believed that the syndrome was a form of en-cephalopathy secondary to a focal or multifocal brainlesions and that the marked sleep activation of the par-oxysmal activity was due to a particularly active syn-chronizing system during slow-wave sleep.

The earlier terms “subclinical electrical status epilep-ticus” and “electrical status epilepticus during sleep”were eventually changed to “continuous spike and waveduring slow sleep” (CSWS) because clinical seizures didnot accompany spike-and-wave discharges in the EEG.Tassinari et al. (1985) thought that the persistent contin-uous spike-and-wave discharges over years were respon-sible for the complex and severe neurologic impairmentsthat developed in the patients.

NOSOLOGY

Although there are similarities, LKS and CSWS arenot presently considered as interchangeable disorders.Landau-Kleffner syndrome is a functional disorder ofchildhood characterized by the following: (1) seizuresthat are relatively easy to treat and self-limited, (2)acquired aphasia, (3) an EEG showing epileptiform dis-charges, usually over one or both temporal regions, and(4) no definitive brain pathology that can explain thebehavioral symptoms and some degree of improvement

when the epileptic condition resolves (Deonna and Rou-let, 1995; Morrell, 1995; Morrell et al., 1995).

One of the terms that have been challenged in recentreports is “acquired aphasia.” Rapin et al. (1977) haveargued that the condition is not aphasia but rather averbal agnosia. Even more recent evidence implies that itmight be an auditory agnosia (Morrell et al., 1995). Thedifference in terminology stems from the fact that theterm acquired aphasia suggests a demonstrable age-appropriate language before the onset of paroxysmalEEG and/or seizures. It is possible that an early onset ofthe condition could prevent language development, butthe strict diagnosis of LKS is not possible in a child inwhom language never developed, unless the aphasia isreversed by halting the epileptic process.

The EEG that shows predominantly bitemporal abnor-malities is markedly activated in slow-wave sleep, whenit may show continuous spike-wave discharges, persist-ing for an extended period of time (Morrell, 1995).Clinical seizures are not always reported or can bepresent in a very subtle form (Deonna and Roulet, 1995;Morrell et al., 1995).

All patients improve in language function when theactive phase of spike-and-wave discharges dissipates, butpermanent language deficits and other neuropsychologicsequelae persist, particularly if the paroxysmal EEG hasbegun early in childhood (Bishop, 1985) and continuesduring the critical period of language development (Mor-rell, 1995; Morrell et al., 1995).

Continuous spike and wave in slow-wave sleep is alsoa functional childhood disorder. Its characteristics in-clude (1) a severe paroxysmal EEG disturbance withspike waves occupying �85% of the sleep recording, (2)self-limited clinical seizures, (3) behavioral and cogni-tive deterioration with or without premorbid develop-mental disturbances, and (4) no brain pathology suffi-cient enough to explain this behavioral deterioration andimprovement once the epileptiform activity is stopped oris resolved (Bureau, 1995a; Tassinari et al., 1992).

The above-mentioned characteristics are not, however,totally agreed upon by all. Tassinari et al. (1992) requirea sleep index of 85%, which 85% or more of slow-wavesleep shows continuous paroxysmal activity in the EEG.However, the ILAE (1989) classification of epilepsiesand epileptic syndrome does not require it. There isvariability in the amount of paroxysmal EEG activity ifpatients are followed long enough. Seizures, as in LKS,may also be subtle and go undetected and underreported.However, the seizures are more common and difficult totreat compared with those in LKS. The common anti-convulsant drug therapies usually successfully treat the

463LKS AND CSWS DURING SLOW-WAVE SLEEP

J Clin Neurophysiol, Vol. 20, No. 6, 2003

clinical seizure but do not necessarily eliminate theparoxysmal EEG abnormality.

By the midteen years, there is spontaneous resolutionof the epileptiform discharges, and often the behavioraland/or neuropsychologic deficits stabilize or improve.However, as in LKS, there are behavioral, cognitive, andpermanent language sequelae; the severity depends onthe age of onset and the duration of the active phase ofparoxysmal activity.

UNDERLYING CAUSE AND BASICMECHANISMS OF ACTION

The syndromes of LKS and CSWS are considered tohave a common pathophysiologic mechanism (De Negri,1995; Gordon, 1997; Hirsch et al., 1990; Maquet et al.,1995; Morrell, 1995; Morrell et al., 1995; Rossi et al.,1999; Tassinari et al., 2000). As noted earlier, bothfunctional disorders begin during the time of corticalsynaptogenesis when the elemental functional circuitry isbeing established in the brain—between the ages of 1and 8 years. In the process of synaptogenesis there isabundant axonal sprouting that results in a doubling inthe number of axonal process and synaptic contact (Hut-tenlocher et al., 1982; Huttenlocher and de Courten,1987; Purves, 1988; Purves and Lichtman, 1980). Themajor factors that determine which of these synapses willbe strengthened and which will be discarded or “pruned”are the neuronal activity or synaptic use (Huttenlocherand de Courten, 1987; Purves, 1988). More than thegenetic programming, environment is the crucial role-player in the establishment of permanent synaptic con-tacts. If a paroxysmal disturbance of a great magnitudedevelops during this time of high axonal sprouting andsynapse formation, the epileptic activity can act tostrengthen synaptic contacts that normally would bepruned for the neuronal aggregates to mediate normalbehavior (Morrell et al., 1995). In LKS, the paroxysmalactivity reinforces these contacts in the developing tem-poroparietal cortex, producing the permanent languagedysfunction (Morrell et al., 1995). The disturbance musthave a bilateral effect to prevent the transfer of functionto the contralateral homotopic cortex. In CSWS, the mostprominent paroxysmal activity appears to be in the fron-tal area and the prefrontal cortex, which disrupts thehigher cognitive and executive functioning before dam-aging language function.

One of the major difficulties in our understanding ofthe spectrum of these age-related encephalopathies is thelow incidence of both LKS and CSWS. One hundredthree cases were reported at the Venice colloquium(Beaumanoir et al., 1995), and from that number, 71

seemed to fit the criteria of CSWS and 31 were LKS.One case remained unclassified. The overall data suggesta definite overlap between the two syndromes. Differ-ences seen in the seizure type, clinical presentations, andthe sequelae—both neuropsychologic and behavioral—could be ascribed primarily to the initial cortical areainvolved, age of onset of the EEG abnormalities relativeto age-dependent synaptogenesis, length of time theparoxysmal EEG was present, and persistence of thecontinuous spike-and-wave discharges during sleep. It isthe conclusion of this study that LKS is a subtype ofCSWS.

Because the functional disruptions caused by thespread of the paroxysmal EEG activity can encompasslarge cortical regions, the two clinical syndrome disor-ders tend to become more similar as the diseaseprogresses. They may produce severe cognitive, behav-ioral, and, ultimately, social dysfunction, mimicking achild with severe autism. The proposed pathophysiologicmechanism also explains that children who are at great-est risk for permanent and irreversible damage are thosewith early onset of the disorder and a prolonged activephase.

The underlying cause may be any pathology capableof generating an epileptic condition. The literature con-tains examples of neuronal migration abnormalities, en-cephalitis, vasculitis, subpial gliosis, and cysticercosis(Cole et al., 1988), all being associated with either LKSor CSWS. In one surgical series of 14 patients with LKS,13 patients in whom biopsy specimens were taken fromthe temporal pole revealed some form of pathologicabnormality (Smith et al., 1992).

EPIDEMIOLOGY

The incidence of LKS cannot be estimated accuratelyat this time. If one uses the strict narrow diagnosticcriteria, LKS is indeed a rare disorder of children. Since1957, 198 cases have been published; 81 cases werereported between 1957 and 1980, but 117 were reportedin the decade between 1980 and 1990 (Beaumanoir,1995). In a Paris psychiatric clinic, Dugas and his col-leagues (1976) reported one new case each year.

The frequency of CSWS is also uncertain. When usingthe strict criteria that include the continuous spike-and-wave discharge occurring during 85% of sleep and rec-ognizable cognitive and behavioral decline, it is also arare disorder. Between 1971 and 1984, there were 19cases reported by Tassinari et al. (1985) and an addi-tional 25 from the medical literature. Since 1984, 10 newcases have been seen at the Center St. Paul, a rate of oneto two per year (Tassinari et al., 1992). Boys are affected

464 M. C. SMITH AND T. J. HOEPPNER


more commonly than girls, with the peak onset occurringbetween 5 and 7 years of age (Bureau, 1995b).

CLINICAL PRESENTATION

Landau-Kleffner Syndrome

The disorder typically presents with speech distur-bance between the ages of 3 and 8 years in a child whohas already developed age-appropriate language produc-tion. The onset can be subacute, steady, or stuttering andinitially consists of a loss of understanding of spokenlanguage. Eventually, and sooner rather than later,speech output is disrupted and paraphasias and phono-logic errors begin to appear. In severe instances, the childbecomes entirely mute and does not respond to nonver-bal sounds as well. This is quite noticeable, as the childwho up to this point has acknowledged the ringing of thetelephone, knocking on the door, and noises outside thehome will not respond now to these stimuli. The childwill commonly begin to display hyperactivity and anattention deficit. There is a rare progression to severedisinhibition or psychosis (Beaumanoir, 1995; Morrell,1995; Morrell et al., 1995).

The language disorder is probably a verbal auditoryagnosia, that is, difficulty or loss of verbal comprehen-sion, which may be mistaken as acquired deafness, atleast early in the course of the illness. It is then followedby gradual deterioration also in verbal production and,finally, mutism and failure to respond to nonverbalsounds.

The most common seizures include episodes of eyeblinking or ocular deviation, head drops, and minorautomatisms, sometimes with secondary generalization.They have a variable relation to the language deficit, and20% to 30% do not exhibit behavioral seizures at all(Beaumanoir, 1985). Characteristically, the seizures havea benign course—they respond well to anticonvulsantmedication and usually resolve on their own by themidteen years. Prognosis is not as good for the languagefunction. If the aphasia exists for more than 1 to 2 years,complete linguistic recovery is seldom seen, and suchpatients continue to have a lifelong language dysfunction(Bishop, 1985; Morrell, 1995; Morrell et al., 1995).

EEG is abnormal in most patients with LKS andshows predominantly bilateral temporal (mainly poste-rior temporal) spikes or spike-wave discharges that areactivated by sleep. On prolonged sleep recordings, sub-continuous 1.5- to 5-Hz spike-and-wave discharges maybe seen during slow–wave sleep that disappear or frag-ment during rapid eye movement sleep (Marescaux et al.,1990; Morrell, 1995; Morrell et al., 1995) (Fig. 1). It is

this common EEG feature that links both LKS andCSWS into a single entity. Most patients do not demon-strate a sleep index of 85% characteristic of slow-wavesleep in CSWS. However, it has been thought that thiscondition was met at some point during the course of thedisorder and simply not documented by an EEG record-ing because of sampling error (Morrell, 1995). Back-ground activity during wakefulness is usually normal.With the use of the methohexital suppression test andintracarotid amobarbital and EEG dipole mapping (Mor-rell, 1989; Smith et al., 1989), it can be shown that manypatients have a unilateral primary epileptogenic region(Morrell, 1995; Morrell et al., 1995). This can involveeither side, since the contralateral propagation of parox-ysmal discharges creates bilateral dysfunction that dis-rupts normal language development.

Syndrome of Continuous Spike Waves DuringSlow-Wave Sleep

Onset is usually in the first decade, with mean age ofonset at 4 to 5 years. The EEG during wakefulness mayshow focal or bilaterally synchronous spike-wave dis-charges over the anterior hemisphere and generalizedspike wave (1.5 to 3.0 cps), with the spike-wave indexbeing �25%. There is a highly characteristic activationduring non-REM sleep. Almost continuous generalized1- to 3-cps spike waves are recorded in the EEG duringnon-REM sleep, with a spike-wave index of �85%. Theparoxysmal activity during REM sleep is very similar tothat of wakefulness, with focal or bilateral paroxysmaldischarges over the frontocentral region.

FIG. 1. EEG tracing taken during sleep in a patient with LKS beforetreatment. Such patterns may occupy 80% to 90% of slow-wave sleep.Electrode designations are those of the standard 10 to 20 internationalsystem.



Seizures are usually present although not invariable.They are commonly nocturnal. The seizures are focalmotor, complex partial, absence, and secondary general-ized tonic-clonic or hemiclonic type. Tonic seizures, afrequent component of other frontal lobe syndromes, donot occur. Compared with LKS, the seizures are morefrequent in CSWS, but, like LKS, they tend to respond toanticonvulsant therapy.

Most patients have normal neuropsychologic functionbefore the onset of EEG changes, but some do have priorabnormal development. All children have cognitive de-cline during the period of paroxysmal EEG. Reducedattention span, hyperactivity, abnormal behavior, apha-sias, and apraxias can all occur as the result of prefrontaldysfunction, leading often to extensive cognitive declinedescribed as mental retardation or dementia (Bureau,1995a; Tassinari et al., 1992). Although the seizuresrespond very well to antiepileptic drug therapy, it doesnot improve the paroxysmal discharges or cognitivesigns and symptoms. EEG changes are self-limited andgenerally disappear by the midteen years, but focalspikes may continue in the frontal region. Seizures alsoremit by the midteens with some improvement of neu-ropsychologic status, although recovery to normal is theexception rather than the rule.

Tassinari et al. (1992) identified three types of patientswith CSWS based on their seizure types: (1) The firstgroup had orofacial, generalized tonic-clonic, and myo-clonic seizures in sleep similar to the patients withbenign Rolandic epilepsy. (2) The second group hadunilateral partial, generalized tonic-clonic, and absenceseizures. Seizures were more frequent and occurred evenwhile awake. (3) The third group had unilateral partialand generalized tonic-clonic seizures in sleep and ab-sence of seizures, absence of status, and atonic seizureswith fall while awake.

With respect to the neuropsychologic and psychomo-tor difficulties, Tassinari et al. (1992) identified twosubgroups. The first subgroup had normal developmentbefore the onset of the CSWS. During the active phase ofepileptiform discharges, a severe decline in IQ, signifi-cant reduction in language function, and temporospatialdisorientation occurred in the majority. All exhibitedbehavioral difficulties, which included hyperactivity, ag-gressiveness, disinhibition, reduced attention span, anddifficulty in connecting with the environment. The sec-ond subgroup had preexisting abnormalities of psy-chomotor development. Further deterioration of cogni-tive function and behavior were noticeable but generallynot as prominent from the baseline as that displayed bythe first group.

DIAGNOSTIC EVALUATION

It is most important to document the development andhistory of language production in a child suspected ofhaving LKS or CSWS. All cognitive, behavioral, neuro-psychologic, or school reports are important to establishpremorbid baseline. A suspected child should also un-dergo neuropsychologic testing by a team experienced inboth linguistic and nonlinguistic evaluation. It is impor-tant to determine if specific deficits exist in cognitivefunction because one would expect to see deficits in thelanguage domain in LKS, whereas features of frontaldysfunction would be a hallmark of CSWS.

Testing procedures include a routine EEG and pro-longed video-EEG monitoring or ambulatory monitoring(Beaumanoir, 1995; Morrell, 1995; Morrell et al., 1995).Neuroimaging tests should include structural and func-tional procedures. Computerized amplitude mapping ofEEG, magnetoencephalography, and intracranial EEGrecordings should be considered if surgical therapy isconsidered (Morrell et al., 1995).

Structural neuroimaging is generally normal in LKSbut abnormal in CSWS (Morrell et al., 1995; Tassinari etal., 1992). The abnormalities in CSWS include focalporencephaly, focal pachygyria, diffuse atrophy, andminor white abnormalities, whereas in LKS, the reportedabnormalities are focal pachygyria and mild, diffuseatrophy (Bureau, 1995b).

A small number of patients of LKS or CSWS hadfunctional neuroimaging. When interpreting the results,it is important to note whether radionucleotide injectionwas given and the scan obtained during the active phaseof spike and wave were conducted in the awake or asleepstate. If scans are performed in the awake state, bothsingle-photon emission computerized tomography andpositron emission tomography with 18F-fluorodeoxyglu-cose show an area of decreased blood flow or glucoseutilization. On the other hand, studies done on sedatedpatients with induced continuous spike-and-wave dis-charges show a focal area of increased blood flow orglucose utilization (Maquet et al., 1995; Morrell et al.,1995). Hirsch et al. (1995) reported increased glucosemetabolism during the active phase of spike and waveduring both asleep and awake states. The area of in-creased metabolism was greater in sleep. It was notknown whether the paroxysmal activity was presentduring the awake scan. The hypermetabolism was re-stricted to focal or regional cortical association areas andthe type of neuropsychologic impairment was in goodagreement with the topography of this disturbance. Themetabolic abnormalities of the two syndromes displayedsignificant overlap, suggesting that they represent two



types of the same spectrum of functional disorders ofchildhood (Hirsch et al., 1995; Maquet et al., 1995).Some patients who underwent positron emission tomog-raphy with 18F-fluorodeoxyglucose scans after the res-olution of the active phase of spike-and-wave dischargesshowed persistent hypometabolism in the areas that pre-viously showed hypermetabolism, documenting long-lasting metabolic alterations (Maquet et al., 1995).

Recording of EEG in slow-wave sleep is essential inthe diagnosis of LKS and CSWS. Because of the shortrecording time, a routine EEG study may rarely recordsufficient duration of slow-wave or non-REM sleep.Amitriptyline at a dose of 1 to 2 mg/kg followed by aprolonged EEG recording of 3 hours improves thechance of confirming a significant activation of parox-ysmal abnormalities during slow-wave sleep. The Venicecolloquium provided direct comparison of EEG data inLKS and CSWS. During the active phase of spike-and-wave discharges, more patients with CSWS met thecriteria of a sleep index of 85% than those with LKS. Theaverage frequency of the spike and wave was 2 Hz ineach disorder. Sleep spindles were absent in 10% ofpatients, more frequently in the CSWS group than in theLKS group. If a patient had a frontal spike focus, therewas more likelihood of having a sleep index of �85%,suggesting that intrinsic circuitry and dense callosalconnections of the frontal lobe facilitate generalizationduring sleep. Patients could be divided into two sets,depending on the severity of the EEG abnormality. Thefirst set had a sleep index of �85%, frequently disruptedsleep spindles, only rare focal discharges during sleep,and bursts of spike and wave in wakefulness. This EEGpattern was seen in 70% of patients with CSWS and in40% of those with LKS. The second, less severelyaffected set showed a sleep index between 50% and 80%,often recognizable sleep spindles, more frequent focalslowing during sleep, and absent or rare spike-and-wavedischarges during wakefulness. These EEG findingswere seen in 30% of the CSWS group and in 60% of theLKS group. These data again support a significant over-lap of EEG findings among LKS and CSWS, althoughthose with CSWS are more likely to have a severelyaffected EEG (Beaumanoir et al., 1995). In addition,focal spikes tend to be frontal in CSWS and temporopa-rietal in LKS.

Magnetoencephalography performed on a small num-ber of patients with LKS revealed a focus of slow wavesand epileptiform activity in the posterior temporal areaadjacent to the sylvian fissure, supporting an origin in thedorsal surface of the superior temporal gyrus (Morrell etal., 1995) (Fig. 2). Sobel and colleagues (2000) foundthat 13 of 19 patients with LKS had perisylvian magne-

toencephalographic spikes. Computerized amplitude andpolarity mapping has also been performed in a fewpatients with LKS and CSWS. In LKS, isolated unilat-eral spikes during the methohexital suppression test orthe first spike in a burst of spike and wave display atangential dipole with suprasylvian negativity and infra-sylvian positivity, indicating the origin, again, in thedorsal surface of the superior temporal gyrus (Hoeppneret al., 1992; Maquet et al., 1995; Paetau et al., 1999).Amplitude mapping of spikes during wakefulness inthree cases of CSWS showed two to be predominantlyleft frontal with temporal spread and one right frontalwith frontal and temporal spread. The spikes during sleepappeared to have a more diffuse distribution (Farnarier etal., 1995). Although the number of cases studied islimited, they support differences in cortical areas maxi-mally involved in the two syndromes.

A few children with LKS underwent chronic intracra-nial electrode and intraoperative recordings in the courseof surgical treatment (Morrell et al., 1995). These studiesconfirmed that the common paroxysmal abnormality isin the area of the posterior temporal lobe, often maximalon the superior temporal gyrus (Fig. 3). At times, theepileptogenic region is confirmed within the sylvianfissure near the Heschl gyrus (Fig. 4).

DIFFERENTIAL DIAGNOSIS

The differential diagnosis between LKS and CSWS isdifficult (Table 1). Landau-Kleffner syndrome tends toaffect a slightly younger population, presenting first withlanguage dysfunction and only later with other cognitive

FIG. 2. Dipole mapping of the epileptiform potential shown at themidpoint of the tracings. Electrode designations are those of the 10–20system. All are referred to a common electrode on the chin (RF1). Eventhough this is a black and white reproduction of a colored map (upperright diagram), it is possible to discern a negative potential at C3(black) and a positive potential at T3 and T5 (checked areas of thegray scale to the left of the black region). This map was obtained atthe deepest level of methohexital anesthesia when background EEGactivity had ceased and only the epileptiform spike persisted.



and behavioral deterioration. In CSWS, the affectedchildren tend to be slightly older, presenting with moreglobal neuropsychologic and behavioral deterioration be-fore language dysfunction appears (Bureau, 1995a). Theseverity of the seizures and EEG abnormality is not aspronounced in LKS as in CSWS. The spike-and-wavedischarges are maximal in the centrotemporal or poste-rior temporal region in LKS and in the frontal headregion in CSWS (Beaumanoir et al., 1995).

The disorders most commonly confused with LKS andCSWS are pervasive developmental delay and autism.The most important distinguishing feature is the loss ofpreviously achieved developmental milestones with LKSand CSWS. In contrast to LKS, children with pervasive

developmental delay and autism display abnormal, non-verbal intelligence as well as language dysfunction. Chil-dren with early-onset and persistent CSWS may deteri-orate more globally and begin to mimic autism. Althoughchildren with pervasive developmental delay and autismmay have paroxysmal EEG abnormalities, they do nothave the continuous spike-and-wave discharges duringslow-wave sleep.

Mental retardation from a wide variety of causes canoccasionally be misdiagnosed as LKS or CSWS. Acareful history will usually document that motor andother developmental milestones were not met and thatthe child was affected from birth. Neurologic examina-tion is often abnormal, demonstrating abnormalities inthe motor and other systems. These children may alsodisplay EEG abnormalities reflecting the underlying cor-tical pathology; however, they rarely meet criteria forcontinuous spike and wave during sleep (Deonna andRoulet, 1995).

Developmental dysphasia is another disorder that maymimic LKS. Children with this condition do not developlanguage skills in the usual time frame. They often havea normal neurologic examination and nonverbal intelli-gence. EEG abnormalities are rare, and a continuousspike-and-wave abnormality has not been described (De-onna and Roulet, 1995). If a child presents with devel-opmental dysphasia and shows continuous spike andwave in sleep, early-onset LKS is a distinct possibility.

Once seizures and EEG recordings have indicated thepresence of an epileptic disorder, more common epilepticsyndromes must be excluded. These include benignchildhood epilepsy with centrotemporal spikes (BECT),other idiopathic localization-related epilepsies, and theLennox-Gastaut syndrome (Bureau, 1995a).

The syndrome of continuous spikes and waves duringslow wave sleep and the Lennox-Gastaut syndrome sharesome common features. These include the presence ofatypical absence and atonic seizures. The EEG of theLennox-Gastaut syndrome also shows slow spike andwave that may activate with sleep, but not to the extentseen in CSWS. In addition, unlike CSWS or LKS, theEEG of Lennox-Gastaut syndrome usually displayspolyspike and wave as well as bursts of rhythmic fastactivity (Bureau, 1995a; Tassinari et al., 1992). Finally,children with Lennox-Gastaut syndrome but not withCSWS or LKS have tonic seizures as one of the prom-inent seizure types. Although there are clear clinicaldifferences in these syndromes, a common feature maybe that severe epileptiform activity contributes to a pro-gressive decline in cognitive function.

The idiopathic localization-related epilepsies are moredifficult to distinguish pathophysiologically from LKS

FIG. 3. MRI of a patient with LKS with magnetoencephalographicdipole superimposed. In a saggital section of the left sylvian regionshown, the dipole orientation is perpendicular to the sylvian fissure andis angled in the anterior posterior plane.

FIG. 4. EEG derived from implanted epidural electrodes in a 5-year-old child with LKS. The bilateral spike-and-wave pattern is welldemonstrated. Epidural electrodes were placed symmetrically over thesylvian fissure in each hemisphere



and CSWS, but they are readily separated clinically(Bureau, 1995a). These disorders have less or no effecton cognitive function because the active spike-and-waveactivity is less severe or because it involves different,more “silent” cortical areas. Aicardi and Chevrie (1982)reported a syndrome that displayed active spike-and-wave discharges that became continuous with sleep.These children had no detectable cognitive or intellectualdeterioration. However, the sleep index and the durationof the active phase of spike and wave were not docu-mented. Deonna et al. (1986) had six similar cases butdocumented increasing neuropsychologic dysfunction atthe time of the EEG deterioration. Polypharmacy mayhave worsened the clinical status, because all improvedon tapering antiepileptic medication. It is probable thatthis syndrome is a subset of CSWS, with an older onsetand a shorter course of the active phase of continuousspike and wave. It may also be that more careful neuro-psychologic testing would be able to detect cognitivedeficits (Bureau, 1995b).

Benign childhood epilepsy with centrotemporal spikesis easily distinguished from LKS by the absence ofacquired aphasia; however, BECT may also be a subsetof CSWS. In the three clinical subgroups of CSWSdistinguished by Tassinari et al. (1992), one subgroupwith orofacial and generalized tonic-clonic seizures insleep closely mimics BECT. Absence of seizures has

also been reported with BECT. There are differences,however, in EEG manifestations. In CSWS, focal abnor-malities predominate in the frontal areas, whereas inBECT they are maximal in the centrotemporal regions.There is a clear activation of epileptiform activity inBECT during sleep that at times becomes continuousspike and wave, but the sleep index never reaches 85%(Bureau, 1995a; Caraballo et al., 1989; Massa et al.,2000). There was one reported case of BECT that wors-ened on carbamazepine therapy; this case reached a sleepindex of 85%, had a clinical deterioration with atypicalabsence and falls, but resolved completely with with-drawal of carbamazepine (Caraballo et al., 1989). Mentalretardation and a history of prior neurologic insult arecommon in CSWS but not in BECT. A family history ofepilepsy is reported in 40% with BECT but in only 10%with CSWS (Bureau, 1995a). Despite these clinical dif-ferences, it is probable that a child with early-onset andpersistent BECT with a high sleep index of spike waveswould display cognitive or motor deficits if carefullytested (Deonna, 2000).

TREATMENT AND OUTCOME

With the small number of cases and variable naturalcauses of LKS, no controlled trials of treatment efficacyhave been attempted (Rotenberg and Pearl, 2003). The

TABLE 1. Comparison of Landau-Kleffner syndrome versus the syndrome of continuous spikes and waves

Variable LKS CSWS

Sex 68% male 63% maleAntecedent history 3% encephalopathy 31%; 36% cerebral palsy; 36%

encephalopathyFamily history of epilepsy 3% 10%Age of onset Peak, 5–7; 5% after 9 years of age Peak, 5–7; 20% after 9 years of ageFirst symptom Seizure, 60% Seizure, 80%Second symptom Neuropsychological, 40% Neuropsychological, 40%Seizure types GTC seizure, 35%; unilateral, 26%;

unilateral status, 6%Unilateral, 50%; unilateral status, 6%;

absence, GTC, CPSDuring active phase of spike-and-wave (�) atonic seizure, no significant 1

seizures, clonic/unilateral, CPS� atonic seizure with fall, significant1 in seizures, 1 absence, 1atonic seizure, 1 atypical absence

After active phase of spike-and-wave 19% rare seizure; 81% seizure-free 16% rare seizure, 84% seizure-freeNeuroimaging 13% abnormal 33% abnormalMeet criteria for 85% spike-and-wave

in sleepLess than 50% 78%

Frequency of spike-and-wave 2 Hz 2 HzIctal discharge awake 26% 67%Focal discharges Centrotemporal/parietal, 60% Frontal, 60%Regional predominance of continuous

spike-and-wave during sleepPosterior Anterior

Table adapted from Beaumanoir, 1995 and Carabello et al., 1989.n � 103: CSWS � 71; LKS � 31.LKS, Landau-Kleffner syndrome; CSWS, syndrome of continuous spikes and waves; FH, family history; GTC, generalized tonic-clonic

convulsions; CPS, complex partial seizures.



clinical seizures in LKS and CSWS, as with the otheridiopathic localization-related epilepsies, are, for the ma-jority, not severe and are easy to control (Bureau, 1995b;Maquet et al. 1995; Morrell, 1995; Morrell et al., 1995).One important exception to this is in the subgroup ofCSWS with daily atypical absences and drop attacks(Tassinari et al., 1992). In this group, clinical seizures aresevere and at times are difficult to treat. The clinicalseizures in both LKS and CSWS are self-limited, withonly rare seizures in approximately 20% of both groupsonce the active phase of spike and wave has resolved(Bureau, 1995b).

All available antiepileptic drugs have been used indi-vidually and in combination for the treatment of LKSand CSWS. Efficacy is difficult to ascertain because allantiepileptic drugs effectively control the clinical sei-zures. However, there is little effect on the paroxysmaldischarges in the EEG. As mentioned above, carbamaz-epine may cause a worsening of seizures; hence, itshould be avoided. Valproate alone or in combinationwith a benzodiazepine appears to be the treatment ofchoice (Dulac, 2001).

Corticosteroid therapy or adrenocorticotropic hor-mone appears to have favorable and long-lasting effects(Van Lierde 1995), and some authors suggest that ste-roids should be the treatment of choice, especially innew-onset disease in a young patient (Deonna and Rou-let, 1991; Hirsch et al. 1990; Lerman et al., 1991).Lerman et al. (1991) recommended a high dose for aprolonged period of time (adrenocorticotropic hormone80 IU/d with a 3-month taper; prednisone 60 mg/d witha 3-month taper). They noted that there may be relapseswith steroid reduction and that some children may needto take steroids for months to years. It appears that theearlier the treatment is initiated, the shorter is the dura-tion for which steroid is required and the better is theultimate outcome (Lerman et al., 1991).

The current recommendation is to initially treat withvalproate with or without benzodiazepine. If the epilep-tiform EEG abnormality and cognitive dysfunction per-sists despite high therapeutic antiepileptic drug levels, aseveral-month course of prednisone with careful fol-low-up is indicated.

Morrell and colleagues (1989, 1995) reported on thesurgical treatment by multiple subpial transaction of 14children with LKS (see also Grote et al., 1999). Allchildren had been unable to use language meaningfullyfor more than 2 years and displayed continuous spike-and-wave discharges that were demonstrated to ariseunilaterally in the superior temporal gyrus and surround-ing perisylvian cortex (Fig. 5). With resolution of theparoxysmal EEG abnormality after multiple subpial

transaction, there was marked improvement in languagefunction over time, with 50% recovering age-appropriatelanguage and returning to regular classroom school and29% showing a marked improvement in language func-tion but still undergoing speech therapy. They concludedthat success in recovery of language function dependedon proper selection of patents and resolution of thesevere epileptiform EEG abnormality (Morrell et al.,1995). Other surgical centers have reported similar effi-cacy of multiple subpial transaction in LKS. Presurgicalevaluation requires delineation of an epileptogenic re-gion involving a unilateral posterior temporal lobe thor-ough sophisticated electrophysiologic and neuropharma-cologic tests that include the methohexital suppressiontest, intracarotid amobarbital test, electrical dipole map-ping, and magnetoencephalographic source mapping.The goal in the treatment of LKS and CSWS is thecomplete elimination of the paroxysmal EEG distur-bance, preferably within the first 2 years to preventserious neuropsychologic sequelae (Maquet et al., 1995;Morrell, 1995; Robinson et al., 2001; Tassinari, 1995).

LONG-TERM PROGNOSIS

Long-term prognosis for seizure disorder in both con-ditions is good, with �20% having persistent, usuallyrare, seizures (Bureau, 1995b). However, the long-termneuropsychologic consequences are not benign (Deonnaand Roulet, 1995). The majority of patients who haveeither disorder have some permanent sequelae that limittheir activities. Those with the earliest onset of spike-and-wave discharges and longer persistence of the active

FIG. 5. Electrocorticographic recordings from a pad of electrodesplaced overlying the sylvian fissure. (A) Widespread distribution ofspike discharge observed before multiple subpial transaction. (B) Trac-ings taken immediately after subpial transaction of the planum tempo-rale and the surrounding perisylvian cortex (area denoted by hatchedlines). The relative lack of back rhythms is characteristic of the EEGimmediately after transition. Rhythmic EEG activity returns within 10to 15 minutes.



phase of CSWS have the worse neuropsychologic se-quelae (Rossi et al., 1999).

In 1980, Mantovani and Landau reviewed the long-term prognosis of 9 patients with LKS followed for 10 to28 years after onset (1980). They discovered that theoverall clinical status and language were normal in�50%. Other studies also report that some form ofaphasia persists in the majority (Beaumanoir, 1995; De-onna et al., 1989; Dugas et al., 1976). Only half of thepatients with a history of LKS are able to live a normallife (Maquet et al., 1995; Morrell et al., 1995). Theprognosis for language is usually dependent on the age ofonset and the severity of the epileptic disturbance, thebilateral anatomic location, and the length of the activephase (Deonna, 1991; Hirsch et al. 1995; Maquet et al.,1995; Robinson et al., 2001). These factors also deter-mine the severity of other neuropsychologic sequelae.

The long-term outcome of patients with CSWS is alsopoor (Morikawa et al. 1992). Like LKS, the age of onsetand duration of the active phase of spike-and-wavedischarges seem to be the two characteristics that corre-late with persistent sequelae. Although there is a globalimprovement in all intellectual areas after resolution ofCSWS, complete restoration of function, especially inverbal ability and attention, is rare (Morikawa et al.,1995). Persistent sequelae include a short attention span,hyperactivity, affective symptoms, and language dys-functions, as well as intellectual impairment.

REFERENCES

Aicardi J, Chevrie JJ. Atypical benign partial epilepsy of childhood.Dev Med Child Neurol 1982;24:281—92.

Beaumanoir A. About continuous or subcontinuous spike-wave activityduring wakefulness: electroclinical correlations. In: BeaumanoirA, Bureau M, Deonna T, Mira L, Tassinari CA, eds. Continuousspikes and waves during slow sleep: electrical status epilepticusduring slow sleep. London: John Libbey, 1995:115—8.

Beaumanoir A. The Landau-Kleffner syndrome. In: Roger J, Dravet C,Bureau M, Dreifuss FE, Wolf P, eds. Epileptic syndromes ininfancy, childhood and adolescence. London: John Libbey, 1985:181—91.

Beaumanoir A, Bureau M, Deonna T, Mira L, Tassinari CA, eds.Continuous spikes and waves during slow sleep: electrical statusepilepticus during slow sleep. London: John Libbey, 1995.

Bishop DVM. Age of onset and outcome in “acquired aphasia withconvulsive disorder” (Landau-Kleffner syndrome). Dev MedChild Neurol 1985;27:705—12.

Bureau M. Continuous spikes and waves during slow sleep (CSWS):definition of the syndrome. In: Beaumanoir A, Bureau M, DeonnaT, Mira L, Tassinari CA, eds. Continuous spikes and waves duringslow sleep: Electrical status epilepticus during slow sleep. Lon-don: John Libbey, 1995a;17—26.

Bureau M. Outstanding cases of CSWS and LKS analysis of the datasheets provided by the participants. In: Beaumanoir A, Bureau M,Deonna T, Mira L, Tassinari CA, eds. Continuous spikes andwaves during slow sleep: electrical status epilepticus during slowsleep. London: John Libbey, 1995b;213—6.

Caraballo R, Fontana E, Michelizza B, Zullini E, Sgro ÁV, Pajno-Ferrara F, Dalla Bernardina. Carbamazepine, assenze attipiche,crisi atoniche e stato di PO continua del sono (POCS). Boll LegaIt EpiI 1989;66/67:379—81.

Cole AJ, Anderrnann F, Taylor L, Olivier A, Rasmussen T, RobitailleY, Spire J-P. The Landau Kleffner syndrome of acquired epilepticaphasia: unusual clinical outcome, surgical experience, and ab-sence of encephalitis. Neurology 1988;38:31—8.

Commission on Classification and Terminology of the InternationalLeague Against Epilepsy. Proposal for revised classification ofepilepsies and epileptic syndromes. Epilepsia 1989;30:389—99.

De Negri M. The maturational development of the child: developmen-tal disorders and epilepsy. In: Beaumanoir A, Bureau M, DeonnaT, Mira L, Tassinari CA, eds. Continuous spikes and waves duringslow sleep: electrical status epilepticus during slow sleep. Lon-don: John Libbey, 1995:3—8.

Deonna T. Acquired epileptiform aphasia in children (Landau-Kleffnersyndrome). J Clin Neurophysiol 1991;8:288—98.

Deonna T. Rolandic epilepsy: neuropsychology of the active epilepsyphase [Review]. Epileptic Disord 2000;2(suppl 1):S59—61.

Deonna T, Peter C, Ziegler A-L. Adult follow-up of the acquiredaphasia-epilepsy syndrome in childhood: report of 7 cases. Neu-ropediatrics 1989;20:132—8.

Deonna T, Roulet E. Acquired epileptic aphasia (AEA): definition ofthe syndrome and current problems. In: Beaumanoir A, Bureau M,Deonna T, Mira L and Tassinari CA, eds. Continuous spikes andwaves during slow sleep: electrical status epilepticus during slowsleep. London: John Libbey, 1995:37—45.

Deonna T, Roulet E. Epilepsy and language disorder in children. In:Fukuyarna Y, Kamoshita S, Ohtsuka C, Suzuki Y, eds. Modernperspectives of child neurology. Tokyo: The Japanese Society ofChild Neurology, 1991:259—66.

Deonna T, Ziegler AL, Despland PA. Combined myoclonic-astatic andbenign focal epilepsy of childhood (atypical partial epilepsy ofchildhood): a separate syndrome? Neuropediatrics 1986;17:144—51.

Dugas M, Grenet P, Masson M, Mialet JP, Jaquet G. Aphasie del’enfant avec epilepsie: Evolution regressive sous traitement anti-epileptique. Rev Neurol (Paris) 1976;132:489—93.

Dugas M, Masson M, Le Heuzey MF, Regnir N. Aphasie. “acquise” del’enfant avec epilepsie (syndrome de Landau et Kleffner): Douzeobservations personnelles. Rev Neurol (Paris) 1982;138:755—80.

Dulac O. Epileptic encephalopathy. Epilepsia 2001;42(suppl 3):23—6.Farnarier, G, Kouna P, Genton P. Amplitude EEG mapping in three

cases of CSWS. In: Beaurnanoir A, Bureau M, Deonna T, Mira L,Tassinari CA, eds. Continuous spikes and waves during slowsleep: electrical status epilepticus during slow sleep. London:John Libbey, 1995;91—8.

Gordon N. The Landau-Kleffner syndrome: increased understanding[review]. Brain Dev 1997;19:311—6.

Grote CL, Van Slyke P, Hoeppner J. Language outcome followingmultiple subpial transection for Landau-Kleffner syndrome. Brain1999;111:561—6.

Hirsch E, Maquet P, Metz-Lutz M-N, Motte J, Finck S, Marescaux C.The eponym “Landau-Kleffner syndrome” should not be restrictedto childhood-acquired aphasia with epilepsy. In: Beaumanoir A,Bureau M, Deonna T, Mira L, Tassinari CA, eds. Continuousspikes and during slow sleep: electrical status epilepticus duringslow sleep. London: John Libbey, 1995:57—62.

Hirsch E, Marescaux C, Maquet P, Metz-Lutz MN, Kiesmann M,Salmon E, Franck G, Kurtz D. Landau-Kleffner syndrome: aclinical and EEG study of five cases. Epilepsia 1990;31:756—67.

Hoeppner TJ, Morrell F, Smith MC, Chez M, Hasegawa H. TheLandau-Kleffner syndrome: a peri-sylvian epilepsy [abstract].Epilepsia 1992;33(suppl 3):122.

Huttenlocher PR, de Courten C. The development of synapses in striatecortex of man. Hum Neurobiol 1987;6:1—9.

Huttenlocher PR, de Courten C, Garey LJ, Van der Loos H. Synapto-genesis in the human visual cortex: evidence for synapse elimi-



nation during normal development. Neurosci Lett 1982;33:247—52.

Landau WM, Kleffner FR. Syndrome of acquired aphasia with con-vulsive disorder in children. Neurology 1957;7:523—30.

Lerman P, Lerman-Sagie T, Kivity S. Effect of early corticosteroidtherapy for Landau-Kleffner syndrome: case reports. Dev MedChild Neurol 1991;33:257—66.

Mantovani JF, Landau WM. Acquired aphasia with convulsive disor-der: course and prognosis. Neurology 1080;30:524—9.

Maquet P, Hirsch E, Metz-Lutz N, Motte J, Dive D, Marescaux C,Franck G. Regional cerebral glucose metabolism in children withdeterioration of one or more cognitive functions and continuousspike-and-wave discharges during sleep. Brain 1995;118:1497—520.

Marescaux C, Hirsch E, Finck S, Maquet P, Schlumberger E, Sellal F,et al. Landau-Kleffner syndrome: a pharmacologic study of fivecases. Epilepsia 1990;31:768—77.

Massa R, deSaint-Martin A, Hirsch E, Marescaux C, Motte J. Seeg-muller C, Kleitz C, Metz-Lutz M. Landau-Kleffner syndrome:sleep EEG characteristics at onset. Clin Neurophysiol 2000;111(suppl 2):S87—93.

Morikawa T, Seino M, Watanabe M. Long-term outcome of CSWSsyndrome. In: Beaumanoir A, Bureau M, Deonna T, Mira L,Tassinari CA, eds. Continuous spikes and waves during slowsleep: electrical status epilepticus during slow sleep. London:John Libbey, 1995:27—36.

Morikawa T, Seino T, Vagi K. Long-term outcome of four childrenwith continuous spike-waves during sleep. In: Roger J, Bureau M,Dravet C, Dreifuss FE, Perret A, Wolf P, eds. Epileptic syndromesin infancy, childhood and adolescence. 2nd ed. London: JohnLibbey, 1992:257—65.

Morrell F. Electrophysiology of CSWS in Landau-Kleffner syndrome.In: Beaumanoir A, Bureau M, Deonna T, Mira L, Tassinari CA,eds. Continuous spikes and waves during slow sleep: electricalstatus epilepticus during slow sleep. London: John Libbey, 1995:77—90.

Morrell F. Varieties of human secondary epileptogenesis [review].J Clin Neurophysiol 1989;6:227—75.

Morrell F, Whisler WW, Bleck TP. Multiple subpial transection: a newapproach to the surgical treatment of focal epilepsy. J Neurosurg1989;70:231—9.

Morrell F, Whisler WW, Smith MC, Hoeppner TJ, de Toledo-MorrellL, Pierre-Louis SIC, Kanner AM, Buelow JM, Ristanovic R,Bergen D, Chez M, Hasegawa H. Landau-Kleffner syndrome:treatment with subpial intracortical transection. Brain 1995;118:1529—46.

Paetau R, Granstrom ML, Blomstedt G, Jousmake V, Korkman M,Liukkonen E. Magnetoencephalography in presurgical evaluationof children with the Landau-Kleffner syndrome. Epilepsia 1999;40:326—35.

Patry G, Lyagoubi S, Tassinari CA. Subclinical “electrical statusepilepticus” induced by sleep in children. Arch Neurol 1971;24:242—52.

Purves D. Body and brain: a trophic theory of neural connections.Cambridge: Harvard University Press, 1988.

Purves D, Lichtman JW. Elimination of synapses in the developingnervous system. Science 1980;210:153—7.

Rapin I, Mattis S, Rowan AJ, Golden GG. Verbal auditory agnosia inchildren. Dev Med Child Neurol 1977;19:192—207.

Robinson RO, Baird G, Robinson G, Simonoff E. Landau-Kleffnersyndrome: course and correlates with outcome. Dev Med ChildNeurol 2001;43:243—7.

Rossi PG, Parmeggiani A, Posar A, Scaduto MC, Chiodo S, Vatti G.Landau-Kleffner syndrome (LKS): long-term follow-up and linkswith electrical status epilepticus during sleep (ESES). Brain Dev1999;21:90—8.

Rotenberg J, Pearl PL. Landau-Kleffner syndrome. Arch Neurol 2003;60:1019—21.

Smith MC, Pierre-Louis SJC, Kanner AM, Morrell F, Chez M, Hase-gawa H, Whisler W, Clase R, Cochran EJ. Pathological spectrumof acquired epileptic aphasia of childhood. Epilepsia 1992;33(suppl):115.

Smith MC, Whisler W, Morrell F. Neurology of epilepsy. SeminNeurol 1989;9:231—48.

Sobel DF, Aung M, Otsubo H, Smith MC. Magnetoencephalography inchildren with Landau-Kleffner syndrome and acquired epilepticaphasia. Am J Neuroradiol 2000;21:301—7.

Tassinari CA. The problems of “continuous spikes and waves duringslow sleep” or “electrical status epilepticus during slow sleep”today. In: Beaumanoir A, Bureau M, Deonna T, Mira L, TassinariCA, eds. Continuous spikes and waves during slow sleep: elec-trical status epilepticus during slow sleep. London: John Libbey,1995:251—5.

Tassinari CA, Bureau M, Dravet C, Dalla Bemardina B, Roger J.Epilepsy with continuous spikes and waves during slow sleep. In:Roger J, Dravet C, Bureau M, Dreifuss FE, Wolf P, eds. Epilepticsyndromes in infancy, childhood and adolescence. London: JohnLibbey, 1985:194—204.

Tassinari CA, Bureau M, Dravet C, Dalla Bemardina B, Roger J.Epilepsie avec pointes-ondes continués pendant Ie sommeil lent–antérieurement décrite sous Ie nom d’ESES (épilepsie avec état demal électroencephalographique pendant le sommeil lent). In:Roger J, Dravet C, Dreifuss FE, Perret A, Wolf P, eds. LesSyndromes Épileptiques de l’Enfant et de l’Adolescent. 2nd ed.Paris: John Libbey Eurotext, 1992:245—56.

Tassinari CA, Rubboli G, Volpi L, Meletti S, d’Orsi G, Franca M,Sabetta AR, Riguzzi P, Gardella E, Zaniboni A, Michelucci R.Encephalopathy with electrical status epilepticus during slowsleep or ESES syndrome including the acquired aphasia [review].Clin Neurophysiol 2000;111(suppl 2):S94—102.

Van Lierde A. Therapeutic data. In: Beaumanoir A, Bureau M, DeonnaT, Mira L, Tassinari CA, eds. Continuous spikes and waves duringslow sleep: electrical status epilepticus during slow sleep. Lon-don: John Libbey, 1995:225—7.

World Health Organization. The ICD-10 classification of mental andbehavioral disorders: clinical descriptions and guidelines. Ge-neva: WHO, 1992.



l kleffner ingles

Documents