Childhood absence epilepsy

INTRODUCTION — Childhood absence epilepsy (CAE) is a common generalized epilepsy syndrome with a presumed polygenic cause, characterized by typical absence seizures appearing in otherwise healthy school-aged children. CAE is one of the most common forms of pediatric epilepsy.

The clinical features, diagnosis, treatment, and prognosis of CAE are discussed here. Other epilepsy syndromes with an onset in childhood are discussed separately. (See "Seizures and epilepsy in children: Classification, etiology, and clinical features" and "Epilepsy syndromes in children" and "Juvenile myoclonic epilepsy".)

CLASSIFICATION — CAE is classified as an epilepsy syndrome with presumed polygenic cause according to the International League Against Epilepsy (ILAE) [1-3]. (See "ILAE classification of seizures and epilepsy".)

●CAE is one of the idiopathic generalized epilepsies (IGEs) – The ILAE described a broad group of genetic generalized epilepsies (GGEs) that are characterized by generalized seizure types and generalized spike-wave with a presumed genetic etiology (figure 1). Contained within the GGEs is a subgroup of IGEs comprised of four syndromes [3]:

•CAE

•Juvenile absence epilepsy (JAE)

•Juvenile myoclonic epilepsy (JME)

•Epilepsy with generalized tonic-clonic seizures alone (GTCA)

●Shared features of IGEs – The IGEs are the most common epilepsy syndromes within the GGEs and have several shared characteristics, including [3]:

•A relatively good prognosis for seizure control

•Typically normal development

•Clinical overlap among CAE, JAE, and JME

•Possible evolution with age from one to another IGE syndrome

•Similar electroencephalography (EEG) findings (2.5 to 6 Hz generalized spike-wave that may activate with hyperventilation or photic stimulation)

•Polygenic inheritance likely

ETIOLOGY AND GENETICS — CAE has a strong complex genetic component resulting from variation within multiple genes and their interaction with environmental factors [3]. (See "Principles of complex trait genetics".)

A genetic cause of CAE has been suggested by family and twin studies, although clear single-gene defects that reliably result in a typical CAE phenotype have not yet been identified. Family studies indicate a 17 percent risk of having typical absence seizures in first-degree relatives of patients with CAE [4], and there is a higher concordance rate for CAE in monozygotic than dizygotic pairs [5]. However, most patients with idiopathic generalized epilepsies (IGEs) do not have a family history of epilepsy [3].

●Susceptibility genes – Several candidate regions and genes have been identified as possible susceptibility loci in CAE. Copy number variations and microdeletions in regions such as 8q and 15q have been reported in a minority of patients with IGE syndromes, including some with CAE [6-9]. One group performed sequencing of the 15q candidate region in 380 patients with CAE and identified heterozygous point mutations in NIPA2, a gene that encodes for a selective magnesium transporter, in 3 of 380 patients with CAE and in none of the 400 controls [10].

Most of the genes otherwise reported code for various subunits of gamma-aminobutyric acid A (GABA-A) receptors [11-18], GABA-B receptors [16,19], calcium channels [16,19-24], or chloride channels [25,26], although validation has generally been lacking and ethnic variation may contribute to the heterogeneity in findings [17,27-30].

●SLC2A1 gene – Another gene of interest in CAE is SLC2A1, which encodes the glucose transporter GLUT1. Mutations in SLC2A1 were found in up to 10 to 12 percent of children presenting with absence epilepsy starting before four years or refractory absences [31-35]. By contrast, no SLC2A1 mutations were found in 84 subjects with CAE as defined by the onset of absences after four years, documented electroclinical typical absence seizures with generalized 3 to 4 Hz spike-wave and normal background EEG activity, normal neurodevelopmental examination, and normal brain imaging [36]. (See 'Differential diagnosis' below and "Overview of infantile epilepsy syndromes", section on 'Glut1 deficiency syndrome'.)

PATHOPHYSIOLOGY — Studies in humans and rodent models have implicated a variety of cortical and thalamocortical circuits in the genesis of typical absence seizures [37].

Most data point to prominent involvement of deep layers of the frontal and somatosensory cortex in the generation of slow wave discharges, with rapid spread to other regions of the cortex and secondary invasion of the thalamus [38-41]. The orbital-medial frontal and medial-lateral parietal cortices have been implicated in studies utilizing simultaneous EEG and functional magnetic resonance imaging (fMRI) during typical slow wave discharges [42,43]. In another study, the posterior cortical regions were active at the onset of the slow wave discharge, followed by spread to other areas later in the seizure, including the thalamic medial dorsal nuclei and the thalamostriate network [44].

Animal data indicate that particular cortical and thalamic excitability and coupling conditions must occur to produce the slow wave discharges of typical absence seizures [38]. Enhanced tonic gamma-aminobutyric acid A (GABA-A) receptor inhibition in thalamocortical neurons has been demonstrated to be necessary and sufficient for the expression of typical absence seizures in various pharmacologic and genetic mouse models [45]. Such inhibition is due to increased extracellular GABA concentrations related to dysfunction of GABA transporter GAT1, an astrocytic transporter involved in GABA synaptic uptake [45].

EPIDEMIOLOGY — CAE is one of the most common forms of pediatric epilepsy, accounting for 12 to 18 percent of epilepsy diagnoses in school-aged children [3,46,47].

In a retrospective analysis of EEG findings of Swedish children aged 0 to 15 years, the annual incidence of "absence epilepsy" was 6.3 per 100,000 [48]. In another report based on a questionnaire sent to patients with seizures in southwest France, the annual incidence of CAE was estimated at 8 per 100,000 [49]. Of note, estimates of the annual incidence of CAE are derived from studies performed in primarily White populations.

Unlike most epilepsy syndromes, CAE is more common in girls (60 to 75 percent of cases) than boys [3,50,51].

CLINICAL FEATURES

Clinical presentation — CAE is characterized by typical absence seizures occurring in otherwise healthy children. The median age of onset of CAE is six years [52,53]. The majority of patients present between 4 and 10 years of age [50]. Up to 20 percent of children have a history of febrile seizures and nearly half have a first- or second-degree relative with seizures [52].

Seizure semiology — The hallmark seizure type of CAE is the typical absence seizure. A minority of patients may develop generalized tonic-clonic seizures (GTCS) (see 'Generalized tonic-clonic seizures' below). Other seizure types are rare in CAE.

Typical absence seizures — Typical absence seizures present as short and frequent episodes of profound impairment of consciousness, usually without loss of body tone, lasting approximately 10 seconds. Loss of body tone or incontinence can sometimes occur [3]. They are abrupt in onset and termination and are easily provoked by hyperventilation. These usually come to the attention of parents, caregivers, or teachers after a prolonged period of observation, despite their high frequency. Typical absence seizures may occur multiple times daily, but their subtle clinical symptoms allow them to go unnoticed or misdiagnosed as inattention.

The electroclinical spectrum of typical absence seizures in CAE is illustrated by two studies that reviewed a combined total of over 2000 absence seizures in more than 450 drug-naïve children with a new diagnosis of CAE [3,52,54]:

●Seizure onset – Approximately half of all recorded seizures begin with an arrest in activity, which is typically complete. Less common first signs include eyelid movements, eye opening, and oral automatisms.

●Ictal signs and symptoms – The majority of electrographic seizures (approximately 80 percent) are associated with at least one clinical sign. The three most common seizure features are pause/stare (80 to 95 percent), motor automatisms (40 to 60 percent), and eye involvement (55 percent). Approximately one-third of seizures involve 3 Hz regular eyelid movements. Motor automatisms are predominantly oral. The state of consciousness is deeply disturbed, with lack of awareness during the episode and inability to recall the event.

●Seizure duration – The typical seizure duration is 4 to 20 seconds, with a median duration of 9 to 10 seconds. Approximately 25 percent of seizures last less than four seconds and approximately 10 percent are longer than 20 seconds.

Hyperventilation is an effective trigger of typical absence seizures, provoking seizures in 90 percent of children old enough to perform the maneuver [52]. Typical absence seizures are rarely provoked by sensory or visual stimuli, although intermittent photic stimulation may provoke a photoparoxysmal response in a minority of patients [3].

The presence of high-amplitude limb myoclonias, head nods, hypotonia, or focal signs are not consistent with typical absence seizures and should raise suspicion for other types of seizures, such as atypical absence, myoclonic absence, myoclonic-atonic, or focal seizures. In addition, atypical absences should be suspected if the duration of the events exceeds the classic 10 to 20 seconds of typical absence seizures, especially if onset and termination are progressive rather than abrupt. (See 'Differential diagnosis' below.)

Generalized tonic-clonic seizures — GTCS may be seen in patients with CAE, but only rarely before puberty. The presence of GTCS prior to or during the active period of typical absence seizures is considered an alert criterion that reduces confidence in the diagnosis of CAE. (See 'Diagnosis' below.)

While GTCS have been reported in up to 60 percent of children in some series [50,53], the true proportion is likely much lower when stringent diagnostic criteria for CAE are used. In two long-term follow-up studies of children with CAE, only 12 to 13 percent of children with CAE developed GTCS [55,56]. In the larger series, the median time to development of GTCS was 4.7 years, and the median age at first GTCS was 13 years [56].

Behavioral and psychiatric comorbidities — Attention deficit hyperactivity disorder (ADHD), anxiety, mood disorders, and learning disorders are common comorbidities in children with CAE [3,57-59]. Similar findings have also been reported in children with other idiopathic generalized epilepsy syndromes that are typically associated with a favorable prognosis. In these reports, cognitive, social, emotional, or psychiatric comorbidities frequently antedate the epilepsy onset and require special educational interventions [54,57,60-64].

ADHD — ADHD is a common comorbidity of CAE in many children and is not simply the result of frequent seizures [57,65]. The pattern of deficits tends to be more inattentive than hyperactive, which can make recognition of the deficits difficult for teachers and parents or caregivers [65-67]. Seizure duration rather than overall seizure frequency is one predictor of attentional dysfunction, but additional risk factors are not well established. In a cohort of 445 children with CAE, 36 percent of children had baseline attention deficits despite intact cognitive functioning [67]. Another study of the same cohort found that the rate of subtle attentional difficulties was higher in the 30 percent of children who had at least one seizure longer than 20 seconds on initial EEG [68]. However, the overall number of seizures was not associated with measures of attention or executive function. (See "Attention deficit hyperactivity disorder in children and adolescents: Clinical features and diagnosis".)

Cognition and learning — Children with CAE typically have intact cognition and intellect at the time of diagnosis. Formal neurocognitive testing may detect mild deficits in a variety of domains, particularly executive function [61]. In a study that compared neuropsychologic and academic function in 94 children aged 8 to 18 years with new-onset epilepsy with 72 healthy subjects, specific difficulties in fine motor skills and executive functions were noted in children with both childhood and juvenile absence epilepsy [60].

In another study that examined behavioral, cognitive, and linguistic comorbidities in 69 children with CAE and 103 age-matched healthy subjects, children with CAE had significantly lower scores on global, verbal, and performance intellectual quotients and speech language quotients [57].

Anxiety and depression — Patients with CAE have higher rates of anxiety and depression compared with the general population [57,69]. In a meta-analysis of five case-control studies with 231 patients with absence seizures and 281 controls, the likelihood of developing depression or anxiety was greater for patients with absence seizures compared with controls (29.0 versus 7.5 percent, odds ratio 4.93, 95% CI 2.91-8.35) [69].

EVALUATION — The evaluation of patients with suspected absence seizures includes a history, examination, and EEG, as detailed below. Brain imaging is not necessary for the diagnosis of CAE in otherwise typical cases.

History — The clinical features are important to differentiate CAE from other types of epileptic and nonepileptic events. The history should focus on event description, including frequency and duration, age of onset, history of other seizure types, and developmental history; CAE typically occurs in otherwise healthy school-aged children [70].

Eyewitness descriptions of events from parents or teachers are critical to the diagnosis. Features that suggest absence seizures should be specifically questioned [3,70]:

●Staring spells or sudden loss of facial expression

●Repetitive movements such as eye blinking, head nodding, or lip smacking

●Interruption of activity

●Events unresponsive to external stimuli

●Occurrence of events regardless of setting (eg, occurring during activity such as playing and not only in passive settings such as watching TV)

Examination — Most children with suspected CAE will have a normal physical and neurologic examination. The goal of the examination is to detect signs of an underlying medical or neurologic disorder other than CAE that may be the cause of the events (eg, cognitive impairment, microcephaly, or eye-movement abnormalities suggestive of glucose transporter 1 [GLUT1] deficiency syndrome). (See 'Differential diagnosis' below.)

Hyperventilation — Absence seizures can be provoked by hyperventilation in approximately 90 percent or more of untreated patients with CAE [52,71,72]. Hyperventilation can be performed in the clinic by asking the child to blow hard on a pinwheel, imaginary candle, or other object for three minutes [70]. It is advisable to prepare the child by letting them know you may call out a word during the hyperventilation trial and will ask them to repeat that word back, to check for retained consciousness during a pause in breathing.

Electroencephalography — A routine EEG with hyperventilation is usually sufficient to record an absence seizure and obviate the need for a video-EEG. If the initial EEG is normal but the clinical suspicion for CAE is high, then a sleep-deprived or prolonged video-EEG study that includes both hyperventilation and intermittent photic stimulation (IPS) may be helpful to maximize the chance of recording an absence seizure and elucidate the diagnosis. Careful annotations of the recording regarding the response of the patient to test items are critical.

Among 47 consecutive patients with newly-diagnosed CAE, diagnostic 30-minute sleep-deprived video-EEG recordings captured an average of six seizures per child; of these, 47 percent occurred with hyperventilation, 25 percent during drowsiness, 13 percent during the awake state, 9 percent during IPS, and 7 percent during sleep [52]. At least one seizure was provoked during hyperventilation in 83 percent of children.

●Interictal EEG – The interictal EEG in CAE shows normal background activity, although it is common to see occasional generalized spike-wave discharges or focal abnormalities. In a review 445 pretreatment EEGs, occipital intermittent rhythmic delta activity (OIRDA) was noted in 21 percent, focal sharp waves in 2.5 percent, and focal slowing in 0.7 percent [68]. In another study, focal intermittent paroxysmal activity was noted in 38 percent of 29 children with CAE, consisting in most cases of frontal spike-wave or polyspike and wave discharges [73]. IPS may provoke a photoparoxysmal response (generalized spike-wave) in up to 21 percent of patients [3,74]. (See "Photosensitive epilepsies", section on 'Photoparoxysmal response'.)

●Ictal EEG – The EEG appearance of a typical absence seizure consists of generalized 2.5 to 5.5 Hz spike-wave discharges (classically 3 Hz) with abrupt onset and termination (waveform 1) [52]. Spike-wave discharges are often stereotyped but may begin and end with discharges that have smaller amplitudes than the main body of the ictus. Pre-ictal and postictal slowing is often present [52].

The morphologic characteristics of the spike-wave discharges vary across recordings. In one study, 87 percent of spike-wave discharges contained only one or two spikes per wave during the seizure; four or more spikes per wave were present in 8 percent of the 47 children when analyzed individually [52]. In another study, single spike-wave discharges were noted in 604 of 721 discharges (84 percent), whereas polyspike wave discharges were present in only 3 percent [68].

The ictal EEG of atypical absences usually shows discharges of irregular generalized spike waves of lower frequency (<2.5 Hz) than those observed in typical absence seizures.

In up to 50 percent of cases, the earliest ictal abnormalities are focal [52]. This is often predominant in the frontal EEG electrodes bilaterally, a phenomenon referred to as "frontal lead-in" or "frontal absences" by some authors [75,76]. Similar features have also been reported arising from the occipital lobes [52]. However, an EEG with a focal "lead-in" of absence seizures is not necessarily indicative of a focal structural lesion or focal epilepsy. Brain imaging, seizure semiology, and other EEG features should be considered to determine if a process other than CAE is occurring in those patients who demonstrate seizure intractability.

Neuroimaging — Brain MRI is by definition normal in patients with CAE, and structural imaging is not necessary for the diagnosis if patients have typical clinical and EEG findings.

High-resolution structural imaging has been used to demonstrate subtle anatomical variation in various brain regions in patients with CAE, although the functional and clinical significance of these findings is uncertain. Reported abnormalities include decreased thalamic gray matter volume and subcallosal gyrus atrophy [77], gray matter loss in the left orbital-frontal gyrus and bilateral temporal lobes [78], and amygdala volume loss in CAE patients with symptoms of attention deficit hyperactivity disorder (ADHD) [79]. Volumetric analyses of 3 Tesla MRI scans in 18 patients with CAE and 18 age-matched control subjects showed that patients with CAE had reduced bilateral frontotemporal cortical gray matter volume and increased posterior medial cortical thickness [80].

CSF and genetic testing — To rule out GLUT1 deficiency, lumbar puncture for cerebrospinal fluid (CSF) analysis (looking for low CSF glucose) and genetic testing (for SLC2A1 pathogenic variants) are indicated for children with any of the following [3]:

●Onset at three years of age or younger

●Microcephaly

●Intellectual disability

However, these tests are not needed in most patients. (See 'Exclusionary criteria' below and 'Other epilepsies' below.)

DIAGNOSIS — CAE is diagnosed based on history, examination, and video-EEG findings along with updated diagnostic criteria as proposed by the International League Against Epilepsy (ILAE) in 2022 (table 1) [3,50,81].

Mandatory criteria — A diagnosis of CAE requires [3]:

●Typical absence seizures (see 'Typical absence seizures' above)

●EEG with paroxysms of 3 Hz (range 2.5 to 4.0 Hz) generalized spike-wave discharges at the start of the absence

Alert criteria — The following alert criteria are absent in most patients with CAE. Alerts do not exclude the diagnosis but reduce confidence in diagnostic certainty and should prompt further investigations to rule out other conditions [3]:

●Generalized tonic-clonic seizures (GTCS) prior to or during the period of frequent absence seizures

●Staring spells with typical duration greater than 30 seconds or with postictal confusion or fatigue

●Absences occurring less than daily in an untreated patient

●EEG findings:

•Consistently unilateral epileptiform discharges

•Lack of hyperventilation-activated 2.5 to 4.0 Hz generalized spike-wave in an untreated patient who performs hyperventilation well for three minutes or longer

•Recording a typical staring spell without EEG correlate in a child with a history of 2.5 to 4.0 Hz generalized spike-wave

•Persistent slowing of the EEG background in the absence of sedating medication

●Age at onset 2 to 3 years or 11 to 13 years

●Mild intellectual disability at onset

●Potentially relevant neurologic examination and/or imaging abnormalities, excluding incidental findings

Exclusionary criteria — The following are exclusionary for the diagnosis of CAE [3]:

●Any of the following seizure types:

•Prominent myoclonic seizures

•Prominent eyelid myoclonia

•Myoclonic absence seizures

•Atonic seizures

•Tonic seizures

•Atypical absence seizures

•Focal impaired awareness seizures

●EEG with diffuse background slowing

●Age at onset less than 2 years or greater than 13 years

●Moderate to profound intellectual disability at onset

●Cognitive stagnation or decline

●Low cerebrospinal fluid (CSF) glucose and/or SLC2A1 pathogenic variant

DIFFERENTIAL DIAGNOSIS — The first step in the differential diagnosis of CAE is to identify the nature of the paroxysmal events and determine whether they are electroclinical typical absence seizures or nonepileptic staring spells. This is usually easy to do and often only requires a short video-EEG recording [68].

Nonepileptic disorders — Nonepileptic staring spells can occur in normal children as well as in association with attention deficit hyperactivity disorder (ADHD), autism, and intellectual disability. Unlike typical absence seizures, staring spells can be interrupted by tactile or vocal stimulation, rarely occur during physical activity, and are never associated with automatisms or other motor signs. (See "Nonepileptic paroxysmal disorders in children", section on 'Nonepileptic staring spells'.)

Other epilepsies — In addition to CAE, absence seizures may be seen in other generalized epilepsy syndromes, including juvenile absence epilepsy (JAE) and juvenile myoclonic epilepsy (JME), as well as rare cases of glucose transporter 1 (GLUT1) deficiency syndrome (GLUT1DS), epilepsy with eyelid myoclonia (EEM), and epilepsy with myoclonic absences (EMA):

●Juvenile absence epilepsy – JAE is a generalized epilepsy also characterized by typical absence seizures but with a later peak age of onset of 10 to 12 years. In addition to later age of onset, distinguishing features of JAE include myoclonic seizures in approximately one-fifth of patients and a higher incidence of generalized tonic-clonic seizures (GTCS), which are uncommon in CAE (table 2).

●Juvenile myoclonic epilepsy – Approximately 20 to 40 percent of patients with JME have typical absence seizures, which typically begin several years before the onset of myoclonic seizures and GTCS. Compared with CAE, the typical absence seizures that occur in JME are typically milder with less complete impairment of awareness, and the spike-wave discharges on EEG are often of slightly higher frequency (ie, 4 to 6 Hz). (See "Juvenile myoclonic epilepsy".)

●GLUT1 deficiency syndrome – In patients with early-onset absences (ie, <4 years of age), GLUT1 deficiency syndrome should also be considered [32,82]. GLUT1 deficiency is a genetic disorder characterized by impaired glucose transport across the blood-brain barrier. The diagnosis is suggested by a low cerebrospinal fluid (CSF) glucose level and can be confirmed in most cases by genetic testing for SLC2A1. The phenotypic spectrum includes not only early-onset absence seizures but also other forms of generalized epilepsies, developmental delay, and paroxysmal exertional dyskinesias that usually emerge during childhood or adolescence [35,83-85].

In patients with GLUT1 deficiency syndrome, treatment with the ketogenic diet can result in marked clinical improvement. Other dietary interventions are also under investigation. In small pilot studies, consumption of triheptanoin, a medium-chain triglyceride, has been associated with significant reduction in spike-wave burden, improvement in neuropsychologic performance, and fewer paroxysmal motor events in children with GLUT1 deficiency who are not receiving the ketogenic diet [86,87]. (See "Seizures and epilepsy in children: Clinical and laboratory diagnosis", section on 'Laboratory and genetic testing in undiagnosed epilepsy' and "Ketogenic dietary therapies for the treatment of epilepsy", section on 'Particularly responsive conditions'.)

●Epilepsy with eyelid myoclonia – EEM is a genetic generalized epilepsy with onset in the first two decades of life, characterized by eyelid myoclonia, absence seizures, and photosensitivity [88]. By contrast, prominent eyelid myoclonia is not seen in CAE [89]. (See "Photosensitive epilepsies", section on 'Epilepsy with eyelid myoclonia (Jeavons syndrome)'.)

●Epilepsy with myoclonic absences – EMA is a genetic generalized epilepsy syndrome with daily myoclonic absence seizures as the predominant type [89]. These seizures are associated with 3 Hz upper-limb jerks and progressive arm elevation (ratcheting up), and they typically cause the patient to bend forward during the seizure. By contrast, myoclonic jerks in CAE may occur but are subtle, low amplitude, lack rhythmicity, and are not associated with progressive arm elevation [89].

●Atypical absence seizures – Atypical absences or absences with special features are usually observed in more severe forms of epilepsy, such as Lennox-Gastaut syndrome, as well as epilepsy with eyelid myoclonia and epilepsy with myoclonic absences mentioned above [1]. (See "Lennox-Gastaut syndrome".)

TREATMENT

Goals of therapy — Typical absence seizures are characteristically extremely frequent, occurring multiple times each day. Once they are recognized, they should be treated to improve quality of life, school performance, and social acceptance, and possibly to reduce the risk of (rarely) associated convulsive seizures. Complete seizure freedom can often be attained by pharmacologic treatment.

First-line therapy — For children with newly-diagnosed CAE, we recommend ethosuximide rather than valproic acid or lamotrigine as first-line therapy [90,91].

●Ethosuximide dosing and adverse effects – The typical starting dose of ethosuximide is 5 to 10 mg/kg per day in two divided doses for children younger than six years, and 250 mg twice daily for children six years of age and older. The usual maintenance dose is 15 to 40 mg/kg per day in divided doses. Serum levels of ethosuximide should be checked initially after one to three weeks, with a goal therapeutic concentration of 40 to 100 micrograms/mL (280 to 700 micromol/L). The most common side effects include nausea, vomiting, sleep disturbance, drowsiness, and hyperactivity. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Ethosuximide'.)

●Assessing response – Initial monotherapy should not be abandoned before ensuring that the maximum tolerated dose has been achieved. In most cases, response can be assessed clinically, as typical absence seizures are typically easy to follow once they are recognized. In some cases, EEG can be used to aid in determining the response to therapy if the clinical improvement is unclear after two to four weeks of therapy.

●Supportive evidence – The use of ethosuximide as first-line therapy is supported by results of a randomized, double-blind trial that compared the efficacy and tolerability of ethosuximide, valproate, and lamotrigine in 453 children with CAE. After 16 weeks of treatment, ethosuximide and valproate were significantly more effective than lamotrigine (16-week freedom-from-seizure rates of 53, 58, and 29 percent, respectively), and the rate of drug discontinuation for adverse effects was similar among the three groups [92]. However, valproate was associated with more frequent attentional dysfunction than ethosuximide (49 versus 33 percent) [67,92]. A follow-up study that assessed the same parameters after 12 months of treatment also confirmed that ethosuximide had better efficacy than lamotrigine and fewer side effects than valproate [71].

Patients who fail or do not tolerate first-line therapy — For patients who have inadequate control of typical absence seizures or intolerable side effects on ethosuximide, we suggest switching to valproate monotherapy, although lamotrigine may be a better alternative in females of childbearing age based on the increasingly well-characterized fetal risks of valproate. (See "Risks associated with epilepsy during pregnancy and the postpartum period", section on 'Valproate' and "Risks associated with epilepsy during pregnancy and the postpartum period", section on 'Neurodevelopmental risks'.)

For patients who develop generalized tonic-clonic seizures (GTCS) on ethosuximide, valproate is also an effective second-line therapy, either as an add-on agent or as monotherapy, since it may be effective for both seizure types [90].

Additional options for refractory CAE include topiramate [93], benzodiazepines [50], acetazolamide [50], zonisamide [94], ketogenic dietary therapy, and vagus nerve stimulation. For all of these options, trials that formally evaluate their effectiveness and side effect profile are lacking [95]; however, observational data suggest benefit.

●Valproate dosing and adverse effects – The starting dose of valproate is 15 mg/kg per day in one to three divided doses. The dose is increased by 5 to 10 mg/kg per day at weekly intervals until seizures are controlled or side effects preclude further increases. Daily doses >250 mg should be given in divided doses. The usual maintenance dose is 30 to 60 mg/kg per day in two to three divided doses. A serum valproate level should be checked one to two weeks after the initial dose but can be checked as soon as three to four days after initiation or dose adjustment; the usual range of therapeutic concentrations is 50 to 125 micrograms/mL (346 to 875 micromol/L).

The most common adverse effects of valproate include nausea, vomiting, hair loss, easy bruising, and tremor (table 3A-B). Valproate is also associated with weight gain, obesity, and insulin resistance. Nocturnal enuresis, likely related to impaired sleep efficiency, has been reported in children. Approximately 5 to 10 percent of patients develop alanine aminotransferase (ALT) elevations, usually asymptomatic, during long-term valproate therapy. More serious forms of toxicity include hyperammonemic encephalopathy, acute hepatocellular injury, acute pancreatitis, thrombocytopenia, and teratogenicity. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Valproate'.)

●Lamotrigine dosing and adverse effects – For children and adolescents 2.5 to 13 years of age, the initial dose is 0.3 mg/kg per day in two divided doses for two weeks, then 0.6 mg/kg per day for two weeks; titrated weekly to effect and tolerability to a maximum daily dose of 12 mg/kg per day or 600 mg/day (whichever is less). Multiple titration strategies have been reported. In the trial discussed above (n = 453), the initial lamotrigine dose was 0.3 mg/kg per day for two weeks, followed by 0.6 mg/kg per day for two weeks [92]. Thereafter, the dose was increased at weekly intervals until efficacy or intolerability in 0.6 mg/kg per day increments up to 3 mg/kg per day for one week (week 8), then increased to 4.5 mg/kg per day for two weeks, then 7 mg/kg per day for two weeks, then 9 mg/kg per day for two weeks, and then finally 12 mg/kg per day (week 15); the mean final dose was 9.7±6.3 mg/kg per day.

Neurotoxic adverse effects of lamotrigine are predominantly dizziness and somnolence. Systemic adverse effects include rash and nausea. Rare but potentially life-threatening adverse effects include Stevens-Johnson syndrome, toxic epidermal necrolysis, and angioedema. The risk of serious rash may be increased if lamotrigine is started at a dose that exceeds the recommended initial dose, if dose escalation is too rapid, or if lamotrigine is coadministered with valproate. Lamotrigine is rarely associated with acute multiorgan failure and hypersensitivity reactions, including drug reaction with eosinophilia and systemic symptoms (DRESS), hemophagocytic lymphohistiocytosis, and disseminated intravascular coagulation. Due to its pro-arrhythmic potential, lamotrigine should be avoided in patients who have cardiac conduction disorders. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Lamotrigine'.)

●Evidence supporting second-line therapies – In the randomized trial comparing ethosuximide, valproate, and lamotrigine described above, 208 patients who experienced treatment failure during the double-blind phase were enrolled in an open-label extension study and randomly assigned to second monotherapy with one of the remaining two study drugs [96]. After 16 to 20 weeks of therapy, freedom-from-seizure rates were higher for ethosuximide and valproate (53 and 58 percent, respectively) than for lamotrigine (29 percent). For ethosuximide and valproate, these response rates are comparable with those observed in the first-line setting [92] and support the practice of using second monotherapy rather than add-on therapy in children who do not respond adequately to first-line therapy.

CAE may respond to ketogenic dietary therapy when antiseizure medications are incompletely effective [97,98]. In a case series of 21 patients with childhood and juvenile absence epilepsy, use of the classic ketogenic diet or modified Atkins diet was associated with a >50 percent seizure reduction in 82 percent and complete seizure remission in 19 percent [97]. In the same report, a review of published studies that included 133 patients with absence epilepsy found similar results; 69 percent had a >50 percent seizure reduction and 34 percent became seizure-free. (See "Ketogenic dietary therapies for the treatment of epilepsy".)

Vagus nerve stimulation has been shown to reduce seizure frequency in some patients with CAE and refractory absences [99,100]. (See "Vagus nerve stimulation therapy for the treatment of epilepsy".)

There are mixed reports from small studies with regard to levetiracetam in CAE, with one study suggesting efficacy [101], another reporting aggravation of absences in certain patients [102], and a third study of 72 children showing apparent benefit in 25 percent but a high rate of discontinuation (74 percent) due to poor seizure control and/or side effects [103].

Drugs to avoid — Several antiseizure medications have the potential to aggravate absence seizures in patients with CAE and should be avoided. These include carbamazepine, vigabatrin, gabapentin, and tiagabine. Phenytoin and phenobarbital are known for their ineffectiveness in treating absences and should also be avoided [50].

Duration of therapy — In most cases, seizures respond well to first-line drug therapy and remit before puberty. Antiseizure medication therapy should be continued for a minimum of two years of seizure freedom. After this period, progressive tapering of antiseizure medications can be considered.

PROGNOSIS — CAE is typically responsive to antiseizure medication (see 'Treatment' above). Although often perceived as a benign form of epilepsy, observational studies with long-term follow-up describe a wide range of seizure outcomes in children with CAE. Much of the variability likely relates to differences in diagnostic classification over the years, however, and the inclusion of patients who ultimately go on to have juvenile myoclonic epilepsy (JME) or other syndromes associated with lower rates of seizure remission.

●Seizure-free remission – Seizure-free remission occurs by early adolescence in ≥60 percent of patients with CAE [3,53,55,104,105]. In one study that included 72 children initially diagnosed with CAE, approximately two-thirds of patients were seizure-free at a minimum follow-up of nine years after seizure onset [53]. Seventeen percent were still having seizures but did not take medications. Forty-four percent of patients not in remission had progressed to JME. Significant factors predicting no remission included cognitive difficulties at onset, absence status epilepticus before or during drug treatment, generalized tonic-clonic or myoclonic seizures after treatment onset, abnormal background on initial EEG, and a history of generalized seizures in first-degree relatives.

In another study, over 90 percent of children with CAE diagnosed according to the 1989 International League Against Epilepsy (ILAE) classification were seizure-free after a mean follow-up of 15 years [55]. Those who fulfilled the 2005 criteria had fewer generalized tonic-clonic seizures (GTCS), but their final overall outcome was not significantly different from those who did not. The total duration of epilepsy and the mean age at final remission were 3.9 and 9.5 years, respectively. Epilepsy duration was longer in those children who had become seizure-free more than six months after treatment initiation [55].

●Prolonged refractory epilepsy – A relatively small proportion of children with typical absence seizures are refractory to typical therapeutic approaches for a prolonged period. In one study of 92 such children with refractory typical absence seizures, approximately 50 percent eventually achieved prolonged seizure freedom with or without antiseizure medications [106].

●Evolution to other IGE syndromes – A minority of patients with CAE do not achieve seizure-free remission, and in some of these cases, CAE evolves to another idiopathic generalized epilepsy (IGE) such as JME.

●Persistence of comorbidities – Despite favorable rates of seizure freedom by adolescence, psychiatric and behavioral comorbidities are common in children with CAE and can persist even in the absence of ongoing seizures. Recognition of comorbid diagnoses such as attention deficit hyperactivity disorder (ADHD) is important because effective treatments exist, and ADHD has been associated with decreased health-related quality of life in children with epilepsy. This is discussed in more detail separately. (See "Epilepsy in children: Comorbidities, complications, and outcomes", section on 'Psychiatric and behavioral health'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Seizures and epilepsy in children".)

SUMMARY AND RECOMMENDATIONS

●Genetics – Childhood absence epilepsy (CAE) is an idiopathic generalized epilepsy syndrome (figure 1). A polygenic cause is suggested by family and twin studies; single-gene defects that reliably result in CAE have not been identified. (See 'Etiology and genetics' above.)

●Epidemiology – CAE is a common form of pediatric epilepsy, accounting for 12 to 18 percent of all cases of childhood-onset epilepsy. The prevalence of CAE is higher in girls (60 to 75 percent of cases) than in boys. (See 'Epidemiology' above.)

●Presentation – CAE is characterized by typical absence seizures occurring in otherwise healthy children. The median age of onset is six years; most patients present between the ages of 4 and 10 years. (See 'Clinical presentation' above.)

●Seizure semiology – The typical absence seizure is characterized by abrupt onset and termination, profound impairment of consciousness, and an average duration of 10 seconds. Common features include arrest in activity, staring, repetitive eyelid movements, and automatisms. (See 'Seizure semiology' above.)

●Comorbidities – Most children with CAE have normal development and intellect; however, neurocognitive testing may detect mild deficits, particularly in executive function. Attention deficit hyperactivity disorder, mood disorders, and learning disorders are common comorbidities. (See 'Behavioral and psychiatric comorbidities' above.)

●Evaluation – The evaluation of patients with suspected absence seizures includes (see 'Evaluation' above):

•Diagnostic hyperventilation, which can provoke absence seizures in most untreated patients with CAE (see 'Hyperventilation' above)

•Sleep-deprived video-EEG that includes both intermittent photic stimulation and hyperventilation (see 'Electroencephalography' above)

•Brain imaging is not necessary in otherwise typical cases (see 'Neuroimaging' above)

●EEG – The classic EEG pattern of CAE consists of discrete periods of generalized 3 Hz (range 2.5 to 5.5 Hz) spike-wave discharges that are easily provoked by hyperventilation. (See 'Electroencephalography' above.)

●Diagnosis – The diagnosis of CAE requires typical absence seizures and an EEG with paroxysms of 3 Hz (range 2.5 to 4.0 Hz) generalized spike-wave discharges at the start of the absence seizure (table 1). (See 'Diagnosis' above and 'Mandatory criteria' above.)

Alert criteria are features that are absent in most patients with CAE; their presence reduces confidence in the diagnosis and should prompt further investigations. Examples include absences occurring less than daily in an untreated patient, a variety of atypical EEG findings, and others listed above. (See 'Alert criteria' above.)

Features not compatible with CAE include certain seizure types (eg, prominent myoclonic seizures, atonic seizures, atypical absence seizures, focal impaired awareness seizures), an EEG with diffuse background slowing, age at onset less than 2 years or greater than 13 years, and others listed above. (See 'Exclusionary criteria' above.)

●Differential diagnoses – The differential diagnosis of typical absence seizures in a child includes nonepileptic staring spells, which can be seen in association with normal development, attention deficit hyperactivity disorder, autism, and intellectual disability.

Typical absence seizures may also be seen in other generalized epilepsy syndromes such as juvenile absence epilepsy (table 2) and juvenile myoclonic epilepsy. (See 'Differential diagnosis' above.)

●Treatment

•For children with newly-diagnosed CAE, we recommend ethosuximide rather than valproic acid or lamotrigine (Grade 1B). Ethosuximide is well tolerated and associated with complete seizure freedom in over half of children after 16 weeks of therapy. (See 'Treatment' above.)

•For children who fail or do not tolerate first-line therapy with ethosuximide, we suggest switching to valproate monotherapy (Grade 2B). Lamotrigine is a reasonable alternative in females of childbearing age based on the fetal risks of valproate. (See 'Patients who fail or do not tolerate first-line therapy' above.)

●Prognosis – In most cases, seizures respond well to the first-line drug monotherapy and remit before puberty, without cognitive sequelae. A minority of patients with CAE do not achieve seizure-free remission, and in some of these cases, CAE evolves to another syndrome such as juvenile myoclonic epilepsy. (See 'Prognosis' above.)