EPIDEMIOLOGY

Childhood absence epilepsy is defined as frequent daily absence seizures (pyknolepsy) in normal school age children. It accounts for 8%–15% of all childhood epilepsies¹ with an annual incidence of 4.7–8.0 per 100,000 children between the ages of 1 and 15 years. The average age at presentation is 6 years (range 2–10 years) with previous febrile seizures reported in 11%.²

PATHOPHYSIOLOGY

This syndrome is likely due to complex inheritance involving several genes.³ High concordance rates for absence seizures in monozygotic twins confirm that genes play a major role. However, as the concordance rate is not 100%, nongenetic factors are also involved. Most of the genes known to be associated with CAE encode calcium channel and GABA receptor subunits. Some of these mutations may have major effect but do not explain the phenotypic heterogeneity within families nor are they found commonly in CAE. Mutations in SLC2A1, the gene encoding glucose transporter type 1 (GLUT1), have recently been reported in children with CAE.⁴ Most of these children also had paroxysmal exertional dyskinesia but this was subtle and not diagnosed until a molecular diagnosis was made. GLUT 1 deficiency should be considered in children with family histories suggestive of dominant inheritance of genetic generalized epilepsy (GGE)⁴ as this may have therapeutic implications particularly if the epilepsy is intractable. Microdeletions in 15q13.13 are found in 1%–2% of children with GGE and are felt to behave as a susceptibility component in a polygenic model where a combination of susceptibility alleles contribute to the phenotype in any one patient.³ The presence of this copy number variation increases a person’s risk of developing GGE from 1 in 200 to 1 in 3.³

CLINICAL PRESENTATION

Seizures

Absence seizures are the only seizure at presentation^2,5 occurring up to thousands of times per day. They are enhanced by hyperventilation (HV) in over 90% of children with many clinicians suggesting this is essential to the diagnosis.² The clinical and EEG features of absence seizures are influenced by a number of factors which result in variation from seizure to seizure both between and within individual children. Some of these factors are inherent to the child, such as age, while others, such as state (awake, drowsy, asleep) or provocation (HV, intermittent photic stimulation (IPS)), reflect a changing environment in which the seizures occur.

The mean duration of the absence seizure in CAE is 10 seconds (range 1–44 seconds) with 75% lasting 4–20 seconds.² If response testing is performed most seizures reveal a clinical change.² Children usually arrest or alter their activity if performing a task.² Abnormal eyelid movements are common, with 3-Hz eyelid movements seen in 40% of seizures.² Eye opening occurs in 70% of seizures in which the eyes were initially closed and just over half of children open their eyes in every seizure. Staring is one of the cardinal features occurring in up to 94% of seizures. Occasionally (10% of children) mild myoclonic rhythmic movements of the face and limbs occur during the seizure. Automatisms occur in approximately 22% of spontaneous seizures but occur frequently during seizures recorded in HV. Children who are response tested during a seizure are usually (75%) completely unaware (no response to testing and no memory of event) but occasionally they may show some degree of responsiveness particularly at the end of the seizure.²

DIAGNOSIS

EEG

The EEG in CAE shows interictal fragments of generalized spike and wave (GSW) in 92% of sleep-deprived EEGs (41% while awake, 100% in sleep).^2,6 These fragments have the morphology of GSW but may not have a generalized distribution. Truly focal epileptiform discharges are seen in 15% of children predominantly in the central area.² Interictal polyspikes are seen in 43% of sleep deprived EEGs but only when drowsy or asleep.⁶ Rhythmic posterior bilateral delta activity (PBDA) is seen in 32% of children and varies in frequency between 2.5 and 4 Hz (Fig. 21–1). It is enhanced by hyperventilation and not seen in sleep.² It is notched in 40% of cases and almost always either bilateral or seen independently on both sides.² A photoparoxysmal response (PPR) has been reported in 13%–18% of children with CAE although some investigators consider CAE with photosensitivity a separate electroclinical syndrome.^2,7

The ictal discharge is often not generalized for the first second (Fig. 21–1).² The initial morphology may consist of single or multiple spikes, disorganized or organized spike and wave and is most often in the bifrontal or bioccipital areas.² Although occasionally a unilateral onset can be seen, these children generally also have seizures starting from the other hemisphere as well as seizures which have a generalized onset.² The majority (87%) of seizures have one or two spikes per wave. More than two spikes per wave are most commonly seen in children with a PPR. Eighty percent of seizures consist of organized discharges with complexes of uniform morphology repeated throughout the discharge. However, 20% of the seizures in a sleep deprived EEG show some degree of disorganization, which is more often seen during IPS, drowsiness, and sleep. The frequency of the GSW in CAE ranges from 2.5 to 5 Hz but is usually close to 3 Hz and may be faster in the first second.²

Diagnostic criteria are listed in Table 21–1.

Imaging Studies

Neuroimaging is not necessary in a child who presents with features consistent with the diagnosis of CAE.⁸ If there are atypical findings such as consistent focality on EEG or lack of response to therapy then an MRI should be considered.

Differential Diagnosis

Other epilepsies that present in normal children with only absence seizures include: juvenile absence epilepsy (JAE), eyelid myoclonia with absences, and myoclonic absence epilepsy. JAE is differentiated from CAE by its age of presentation (>10 years) and the frequency of the absence seizures (infrequent and usually not daily). The absences are clinically and electrographically similar to those of CAE although the frequency of the GSW may be slightly faster.⁶ These children may present with generalized tonic clonic seizures (GTCS) with the absences only recognized in retrospect or on the EEG. Seventy-five percent of children develop GTCS, which present at the same time as the absence seizures. This is in contrast to CAE, where if GTCS occur, they usually present after the absence seizures are outgrown.

The rare epilepsy syndrome eyelid myoclonia with absences, or Jeavons syndrome, is not as yet recognized by the International League Against Epilepsy (ILAE). The absence seizures are different to those in CAE in that they are brief and associated with eyelid myoclonia, which consists of 4–6-Hz myoclonic jerks of the eyelids with simultaneous upward deviation of the eyes. These seizures occur mainly on eye closure and absences without eyelid myoclonia do not occur. Children present in early childhood and are photosensitive.⁷ The EEG consists of brief (3–6 seconds) bursts of GSW and polyspikes with a normal background.

TREATMENT

Criteria for Starting Treatment

Children with CAE have significantly more absence seizures per day than are recognized. These frequent seizures, usually with severe loss of awareness, contribute to learning and behavior difficulties as well as accidents.^9,10 Treatment with antiepileptic drugs (AEDs) results in beneficial effects on cognitive functioning in these children and is therefore recommended after an EEG confirms the diagnosis. There is little information in the literature regarding the outcome of untreated CAE as all children are generally treated in the modern era. Older studies before the availability of effective medications lack consistency and certainty of diagnosis as they were prior to both EEG and diagnostic classification schemes. Adie reported in 1924 that children with pyknolepsy aged 4–10 years were not effectively treated with available therapies at that time but that after weeks to years the seizures stopped and there were no residual problems. There is some evidence that longer periods of uncontrolled seizures prior to therapy is associated with more learning and behavioral difficulties. However, it is unclear whether therapy alters prognosis or whether children with CAE that is likely to remit have absence seizures that are easier to treat.

Optimal Treatment Regimen

Psychosocial difficulties as young adults are common following CAE, even in those who outgrow their epilepsy and have discontinued AEDs (Fig. 21–2).⁹ Changes in parenting and expectations may contribute to this. Parents and teachers should be encouraged not to alter the way they interact with their child simply because of the epilepsy. Extra precautions may be necessary due to seizures but the child should not be limited unnecessarily.

Recommended First-Line AEDs—Ethosuximide, Sodium Valproate, and Lamotrigine

Ethosuximide (ETX), sodium valproate (VPA), and lamotrigine (LTG) are all used as first-line medication in CAE.¹¹ The first-line AED used for an individual child should consider efficacy, side-effect profile, and titration schedule of the AED, as well as patient comorbidities (headaches, weight, etc.). The ideal initial AED for one patient may not be ideal for another.

Efficacy

Prior to the 1950s, phenobarbital and trimethadione were used with varying success and considerable side effects. Reasonable efficacy for absence epilepsies was first reported with ethosuximide in 1958 and subsequently a decade later with sodium valproate. Early studies of these drugs included small populations with varying severity of mixed seizure types and epilepsy syndromes.¹² A large observational cohort study of 75 children with CAE reported a positive response to initial monotherapy in 60% (31/50) for ETX and 75% (15/20) for VPA.¹²

Lamotrigine has been shown to be more effective than placebo in a responder enriched open label trial of children (age 2–16) with new onset absence epilepsy.¹³ Twenty-eight children who became seizure free on LTG were subsequently randomized to be blindly weaned to placebo or remain on LTG. Sixty-two percent of the children on LTG remained seizure free compared to 22% of those weaned to placebo.¹³ Subsequently a case series showed LTG eliminated absence seizures in 55% of 20 children with new onset CAE.

Until recently, there has been a lack of strong evidence to support the use of one first-line AED over another in CAE as there were only four randomized controlled trials comparing these AEDs (Table 21–2).^{12,14,15,16,17} These trials had relatively small numbers and failed to show a significant difference between therapies; however, the confidence intervals were wide and important differences could not be excluded.^{12,14,15,17,18} In 2010, a large multicenter double-blind, randomized, controlled trial compared ETX, VPA and LTG in 453 children with CAE.¹⁰ After 16–20 weeks of therapy, VPA and ETX were found to have similar efficacy and be superior to LTG (Table 21–2). The odds ratio with VPA versus ETX was 1.26 (95% confidence interval (CI), 0.8–1.98, p = 0.35), VPA versus LTG was 3.34 (95% CI 2.06–5.45, p = 0.001) and ETX versus LTG was 2.66 (95% CI 1.65–4.28, p = 0.001).¹⁰ Although Coppola and colleagues reported that the efficacy for LTG increased over a 12-month period to approach that of VPA,¹⁵ a recent large observational cohort study of 214 children with new onset GGE (aged 4–16 years) found that children were more likely to stay on VPA (89%) than LTG (69%) at 12 months follow-up due to lack of efficacy of LTG.¹⁹