Clinical Features

Some children with generalized tonic-clonic seizures, or their parents, are aware of the impending seizure, hours or days before it occurs. The child may have a headache, insomnia, irritability, or a change in appetite. This prodrome is to be distinguished from an aura, which is a simple partial seizure that may occur before generalization of the seizure. Unlike the aura, the prodrome is not associated with any EEG epileptiform activity.

As indicated by the name, generalized tonic-clonic seizures have two distinct phases: tonic and clonic [Gastaut and Broughton, 1972]. Loss of consciousness usually occurs simultaneously with the onset of a generalized stiffening of flexor or extensor muscles – the tonic phase. The loss of consciousness usually is complete, and it is rare for patients to have any partial awareness of what happened during the seizure. During the tonic phase, prolonged extension of the back, neck, and all limbs often occurs. The eyes remain open, and a cry or yell is common. The tonic phase typically lasts 10–30 seconds and is followed by the clonic phase, which usually starts with a rapid tremor and then slows to massive jerks of the extremities and trunk. A decrescendo pattern to the frequency (although not necessarily the strength) of the jerks is seen as the seizure ceases. The clonic phase typically lasts 30–60 seconds.

Cyanosis is common and results from the arrest of ventilation during the tonic phase and insufficient short breaths during the following clonic phase. Pupillary dilatation, salivation, sweating, hyperthermia, and incontinence are common.

A postictal state always follows the clonic phase. The duration is quite variable. Some patients, even after a severe and prolonged seizure, respond within a minute or so, whereas other patients may be difficult to arouse for 20–30 minutes. Although the postictal phase initially consists of stupor, it may be followed by confusion or agitation.

In a number of syndromes, generalized seizures are a predominant feature. Epilepsy with so-called grand mal seizures on awakening is a syndrome in which the generalized tonic-clonic seizures occur only on awakening [Betting et al., 2006; Unterberger et al., 2001; Beghi et al., 2006]. In generalized epilepsy with febrile seizures plus, patients initially have febrile seizures; later, as the condition evolves, they experience afebrile generalized tonic-clonic or other types of generalized seizures [Abou-Khalil et al., 2001; Baulac et al., 2004; Scheffer and Berkovic, 1997; Wallace et al., 2001, 2002]. Generalized tonic-clonic or clonic-tonic-clonic seizures often can be a prominent component of juvenile myoclonic epilepsy, a condition discussed in Chapter 56. Children with Lennox–Gastaut syndrome may have generalized tonic-clonic seizures, although tonic seizures are the core seizure type in this syndrome.

Although their disorder currently is not classified as a syndrome, a distinct group of patients with photosensitive epilepsy have seizures provoked by flickering lights and have a photoconvulsive response to strobe light on EEG. The photoconvulsive response is characterized by spike-and-wave and multiple spike-and-wave complexes that are bilaterally synchronous, symmetric, and widespread (Figure 53-1). These should be contrasted with spikes that are time-linked with the photic stimulation and confined to the occipital region, which in children are normal. The photoconvulsive response may be self-limited and cease during stimulation or continue beyond the stimulation [Reilly and Peters, 1973]. The most effective frequency of the flash is 10–20 Hz.

The seizures evoked by photic stimulation usually are primary generalized in type: generalized tonic-clonic, absence, or myoclonic seizures. Although patients with partial seizures may have photoconvulsive responses during an EEG study, it is unusual for a partial seizure to be precipitated by photic stimulation [Jayakar and Chiappa, 1990]. Generalized epileptiform discharges during photic stimulation correlate with generalized seizures, whereas focal discharges are more commonly associated with partial seizures [Gilliam and Chiappa, 1995].

Photosensitive epilepsy can be classified into two major groups: (1) pure photosensitive epilepsy, in which clinical seizures occur only when the patient is exposed to the photic stimulus, and (2) photosensitive epilepsy, in which spontaneous seizures occur in addition to those induced by light stimulation [Binnie and Jeavons, 1992]. Precipitating stimuli that can produce a seizure include sunlight reflected from water, sunlight viewed through leaves of trees in a breeze or while driving along tree-lined streets, discotheque lighting, faulty fluorescent lamp (“nervous ballast”), television, and video games. Photosensitive epilepsy usually appears around puberty, with a mean age at onset of 14 years and a higher incidence in girls than in boys [Jeavons, 1982].

Differential Diagnosis

Difficulty is rarely encountered in correctly diagnosing generalized tonic-clonic seizures (formerly termed grand mal seizures). Generalized tonic-clonic seizures can be divided into those that are primarily generalized from onset and those that begin as partial seizures and then secondarily generalize [Schmidt et al., 1983; Jobst et al., 2001]. Differentiating features are listed in Table 53-1. On very rare occasions, generalized tonic-clonic seizures begin as another generalized seizure type such as absence [Niedermeyer, 1976] before evolving into tonic-clonic activity.

Table 53-1 Comparison of Primary and Secondary Generalized Tonic-Clonic Seizures

	Primary GTC	Secondary GTC
Prodrome	Occasionally	Occasionally
Aura	Never	Frequently
Usual duration	1–3 min	1–3 min
Family history for seizures	Frequently	Occasionally
Acquired lesions	Rarely	Frequently
EEG Interictal Photoconvulsive response Hyperventilation	Normal or generalized spike-wave Frequently May activate spike-wave discharges	Normal or focal or multifocal spikes Rarely Usually does not activate epileptiform discharges

Breath-holding attacks, pseudoseizures (nonepileptic events), and syncope occasionally may resemble generalized tonic-clonic seizures. These are discussed in Chapters 64 and 65.

Electroencephalographic Findings

In patients with partial seizures with secondary generalization, the interictal EEG features are similar to those described for partial seizures. Patients with primary generalized tonic-clonic seizures typically have either normal interictal EEG activity or bursts of generalized spike-and-wave discharges. As with absence seizures, spike-and-wave or polyspike-and-wave activity can occur with photic stimulation and hyperventilation. Normal EEG activity, however, can be observed in patients with generalized tonic-clonic seizures.

The ictal EEG features begin suddenly and are widespread. The tonic phase usually begins as loss of background frequencies, with sudden generalized suppression of the background activity (flattening or attenuation of the EEG signal), followed by a gradual build-up of low-voltage fast spikes, usually starting at 20–40 Hz and then decreasing to 10 Hz, lasting up to 10 seconds, with a progressive increase in amplitude and decrease in frequency – the so-called “epileptic recruiting rhythm.” This is followed by the clonic phase, with slow waves following the spikes. The spike-and-wave complexes gradually slow in frequency before stopping.

During the postictal phase, the EEG shows either suppression of background activity or diffuse slowing. The duration of the postictal changes is variable, with background abnormalities lasting minutes to even days following the seizure. Focal slowing during the postictal phase suggests that the seizure had a partial onset. The absence of any slowing during the postictal phase should raise the question of nonepileptic seizures.

Etiology

The etiology of seizures in children who have secondary generalized tonic-clonic seizures is the same as those described for partial seizures. Patients with primary generalized tonic-clonic seizures usually do not have structural lesions and often have a familial disorder, such as juvenile myoclonic epilepsy. The genetics of generalized tonic-clonic seizures are discussed in Chapter 52.

Initial Evaluation

The initial steps in the evaluation depend to a major degree on the patient’s clinical status on first presentation to the physician. For example, the patient who arrives at the emergency department following a generalized seizure with fever and confusion is evaluated differently from the otherwise normal child who presents several hours after a generalized seizure. The former needs an urgent evaluation to look for an infectious etiology, such as meningitis or encephalitis, whereas the latter can be managed less hastily. This chapter deals with the evaluation and treatment indicated for patients with generalized seizures as their only presenting clinical sign.

In all children with their first seizure, a careful history should be obtained, with particular attention to whether the seizure was preceded by an aura or if the seizure had any focal features. An aura, focal onset to the seizures, or Todd’s paralysis would suggest that the seizure had a partial onset. A review of potential precipitating factors, such as sleep deprivation or photic stimulation, may be useful in counseling the patient and the parents. A past medical history of birth asphyxia or trauma, head injury, prolonged febrile seizures, meningitis, or encephalitis may offer etiologic clues indicating that the seizures have a partial onset. A history of other neurologic symptoms should be sought. Headaches, especially those associated with vomiting or occurring at night, should raise the possibility of a structural lesion.

Although neuroimaging appears to have reduced the importance of the neurologic examination, abnormal findings, even if subtle, may provide clues to the location of seizure onset and etiology of the seizures. For example, asymmetries in facial expression or strength may indicate focal deficits, even in the presence of normal neuroimaging findings.

Diagnostic studies ordered depend on findings during the history and neurologic examination. Children with an unremarkable history other than for the seizures and with normal findings on neurologic examination typically require only an EEG and neuroimaging. An EEG should always be obtained because it may help differentiate partial from generalized seizures. Focal epileptiform activity or slowing would raise the possibility of a structural lesion.

Neuroimaging is recommended in all patients presenting with their first unexplained generalized tonic-clonic seizure, with the recognition that, in patients with normal neurologic examinations, the chance of finding a treatable lesion is quite low. Because generalized tonic-clonic may have a partial onset with subsequent rapid generalization, however, the seizure may be incorrectly diagnosed as primary generalized in type. In nonemergency settings, the imaging test of choice is a magnetic resonance imaging (MRI) scan. Computed tomography (CT) scans can be used in acute situations to determine whether a mass lesion and hemorrhage is present.

The need for further diagnostic testing, such as metabolic screening and cerebrospinal fluid examination, depends on the clinical presentation. In the absence of mental retardation, developmental regression, or abnormalities on neurologic examination, these studies usually are not indicated following the initial seizure.

Comorbidities Associated with Generalized Seizures

While a major goal in the treatment of epilepsy is stopping the seizures, it is not the only treatment goal. Patients with epilepsy are a risk for a number of comorbidities. Comorbidity refers to the co-occurrence of two supposedly separate conditions that occur together more than by chance (also see Chapter 62). Depression occurs more frequently in patients with epilepsy than in the normal population, so that epilepsy and depression are comorbidities. Comorbidities are not necessarily causal. For example, because epilepsy and depression are comorbidities does not mean that epilepsy caused the depression or depression caused the epilepsy. Rather, it is possible that both conditions have a common biological substrate or that another independent variable triggers one of the comorbidities. For example, epilepsy often leads to drug therapy, which could cause depression independently of the epilepsy.

Comorbidities associated with epilepsy include depression [Camfield and Camfield, 2009; Dunn et al., 1999], suicidality [Bridge et al., 2007; Tellez-Zenteno et al., 2007; Donner et al., 2001; McGregor and Wheless, 2006], attention-deficit hyperactivity disorder (ADHD) [Dunn and Kronenberger, 2005], conduct disorders [Dunn et al., 2009], anxiety [Caplan et al., 2005], cognitive impairment and learning disabilities [Fastenau et al., 2004; Aldenkamp et al., 2005], and migraine [Pellock, 2004].

Being aware of the comorbidities frequently associated with childhood epilepsy may influence the choice of antiepileptic drug (AED) used to treat the seizures. For example, lamotrigine and valproate may be helpful in treating both the primary generalized seizures and mood disturbance. Likewise, topiramate and valproate can treat both seizures and migraine. On the other hand, AEDs may exacerbate the comorbid condition. Topiramate can have adverse cognitive effects, while benzodiazepines and barbiturates can exacerbate ADHD and conduct disorders.

Medical Treatment

The medical treatment of generalized seizures that have a partial onset and secondarily generalize is, in general, the same as for the treatment of partial seizures. It is generally assumed that an AED that reduces the likelihood of partial seizures would also prevent partial seizures that secondarily generalize. However, it is possible that there may be a difference in the ability of an AED to prevent secondary generalization following the partial onset.

One of the issues is that, in many studies, partial seizures with secondary generalization are included along with primary generalized tonic-clonic seizures. In the United States, AEDs that have an indication for monotherapy or adjunctive therapy of generalized seizures include phenytoin and carbamazepine. However, while both drugs are effective in partial seizures, they can exacerbate primary generalized seizures [Perucca et al., 1998; Guerrini et al., 1998]. Studies examining efficacy and safety of AEDs in children with primary generalized epilepsy are very limited. Although a number of the newer AEDs are being used for the treatment of primary generalized tonic-clonic seizures, even in adults definitive evidence for their effectiveness is lacking [Faught, 2003; French et al., 2004]. In addition, data for adults and children often are combined when studies are reported. Antiepileptic drug studies done solely in children with primary generalized tonic-clonic seizures are limited. Although most of the newer antiepileptic drugs have been used in the treatment of generalized tonic-clonic seizures, their usefulness in children has not yet been established [Holmes, 2003; Jarrar and Buchhalter, 2003].

The major drugs typically used to treat primary generalized tonic-clonic seizures include valproate [Murphy and Delanty, 2000; Ramsay and DeToledo, 1997; Marson et al., 2007], lamotrigine [Mikati and Holmes, 1997; Trevathan et al., 2006; Marson et al., 2007], levetiracetam [Rosenfeld et al., 2009;Noachtar et al., 2008;Wheless, 2007], and topiramate [Biton et al., 1999; Marson et al., 2007]. Although phenobarbital is effective in the treatment of primary generalized tonic-clonic seizures, the adverse side effects profile has reduced its use as initial therapy for primary generalized tonic-clonic seizures [Taylor et al., 2001].

The four drugs most commonly used for primary generalized tonic-clonic seizures are reviewed briefly here. Additional information about AEDs is available in Chapter 59.

Clinical Features

Absence seizures, formerly termed petit mal seizures, are characterized by an abrupt cessation of activity, change in facial expression, and impairment of consciousness [Browne et al., 1974; Pearl and Holmes, 2008; Porter et al., 1973]. Absence seizures are not common, accounting for less than 10 percent of all seizure types [Hauser and Hersdorffer, 1990; Hauser and Kurland, 1975; Sofijanov, 1982]. Absence seizures may be the most common seizure type to go undetected. The prevalence of absence seizures is highest during the first 10 years of life and then drops dramatically to a very low level [Hauser and Kurland, 1975; Hertoft, 1963; Sato et al., 1976]. Absence seizures are more common in girls than in boys [Hertoft, 1963; Sato et al., 1976]. Typical absence seizures rarely start before the age of 2 years or after the teenage years [Sato et al., 1983; Sato, 1983]. In a study of 83 patients with absence seizures, Sato and co-workers [Sato et al., 1983] found the average age at onset was 3.8 years.

Absence seizures are classified as typical or atypical in type (Table 53-3). Typical absence seizures are short, rarely lasting over 30 seconds; as with other generalized seizures, they are never associated with an aura or postictal impairment. The sudden onset of impaired consciousness, usually associated with a blank facial appearance without other motor or behavioral phenomena, is characteristic. The degree of impairment of consciousness is variable. Some children remember virtually everything that is said during the seizure, whereas for others, the entire duration of the seizure is “time lost.”

Table 53-3 Classification of Absence Seizures

Clinical Seizure Type	EEG – Interictal	EEG – Ictal
Typical absence a. Impairment of consciousness only b. With mild clonic components c. With tonic components d. With automatisms e. With autonomic components

Normal background; short bursts of regular and symmetrical 2.5–4-Hz spike wave; rare frontal or multifocal spikes, sharp waves; photoconvulsive response Regular and symmetrical 2.5–4-Hz spike wave; often induced by hyperventilation or photic stimulation Atypical absence Abnormal background; short bursts of irregular, asymmetrical <2.5-Hz spike wave; multifocal spikes, sharp waves; photoconvulsive response rare Bursts of irregular, asymmetrical <2.5-Hz spike wave; rarely induced by hyperventilation or photic stimulation

Although the absence seizure is commonly thought to consist only of staring, in fact the behavioral changes associated with the seizure type usually are more complex. Most absence seizures are accompanied by motor, behavioral, or autonomic phenomena, and seizures characterized by only staring and altered consciousness are unusual. Penry et al. [1975] reviewed 374 absence seizures recorded on videotape from 48 patients and found simple absences, characterized by staring and cessation of activities, to constitute only 9.4 percent of the seizures. A majority of the patients had other clinical manifestations.

Automatisms, semipurposeful behaviors of which the patient is unaware and subsequently cannot recall, are very common with absence seizures [Holmes et al., 1987; Penry et al., 1975]. They either may be perseverative, reflecting continuation of preictal activities, or may arise de novo. Simple behaviors, such as rubbing the face or hands, licking the lips, chewing, grimacing, scratching, or fumbling with clothes, tend to be de novo automatisms. Complex activities, such as dealing cards, moving a chess piece, or handling a toy, are generally perseverative. Speech, if it occurs during the seizure, usually is perseverative and may be slow and slurred, but also may be totally normal [McKeever et al., 1983]. The longer the absence seizure lasts, the more likely automatisms are to occur [Penry et al., 1975].

Clonic or myoclonic components are common but may be quite subtle, most frequently consisting of blinking. Clonic activity also may be manifested by nystagmus, rapid jerking or trembling of the arms, or head nods. Alterations in muscle tone may lead to stiffening of the trunk or a fall. A study of 426 typical and 500 atypical absence seizures studied in 54 children using simultaneous EEG and video monitoring found myoclonic jerks in 13 percent of typical absences and 12 percent of atypical absences [Holmes et al., 1987]. Children with the syndrome of epilepsy with myoclonic absences have prominent myoclonus during the seizure.

Autonomic phenomena occasionally may be seen with absence seizures and include dilatation of the pupils, pallor, flushing, sweating, salivation, piloerection, and even urinary incontinence.

The frequency of absence seizures varies considerably from day to day and even from hour to hour. The number of absence seizures varies significantly with different environmental situations [Borkowski, et al., 1992; Pearl and Holmes, 2008]. Seizures are more likely to occur during periods of inactivity than when the child is busily engaged in a task. Fatigue also may dramatically increase seizure susceptibility.

A majority of children with typical absence seizures have normal findings on neurologic examination [Dalby, 1969; Sato, 1983]. When neurologic abnormalities are found, they usually are mild and nonprogressive. Most children with typical absence seizures have normal or mildly low intelligence [Dalby, 1969; Sato, 1983]. Compared with age-matched controls, children with absence seizures have lower general cognitive function with impaired visual-spatial skills and memory disturbances [Pavone et al., 2001]. Children with absence seizures have higher rates of behavioral problems and psychopathology than in the normal population [Ott et al., 2001].

Typical absence seizures may be associated with generalized tonic-clonic seizures in 40–60 percent of patients [Hertoft, 1963; Loiseau et al., 1983; Sato, 1983]. In most children, the generalized tonic-clonic seizures occur after onset of the absence seizures [Loiseau et al., 1983].

Although atypical absence seizures form a separate category of absences, overlap between the two seizure types is considerable, and they appear to represent a clinical continuum [Holmes et al., 1987]. Diminished postural tone, or tonic or myoclonic activity, is significantly more likely to be the initial clinical feature in atypical than in typical absences. Automatisms are less likely in atypical absences than in typical absences. Like typical absences, atypical absences have a distinct onset and ending, without auras or postictal symptoms. Although atypical absences usually are of longer duration than typical absences, a considerable amount of variability exists.

Atypical absences usually begin before the age of 5 years and often are associated with other seizure types and mental retardation. In children who have profound mental retardation, it may be difficult to detect subtle behavioral changes associated with the absence seizures. Many children with atypical absence seizures have Lennox–Gastaut syndrome.

Four syndromes are associated with typical absence seizures: childhood absence epilepsy (pyknolepsy), juvenile absence epilepsy, epilepsy with myoclonic absences, and juvenile myoclonic epilepsy.

Differential Diagnosis

The primary diagnostic considerations in the child referred because of “staring attacks” include absence seizures, complex partial seizures, and daydreaming (Table 53-4). Complex partial seizures are more common than absence seizures and also are manifested by an alteration in consciousness with staring, automatisms, changes in tone, and autonomic symptoms. Complex partial seizures tend to be longer and less frequent, but clinically no absolute distinguishing factor may be present. The presence of an aura or postictal impairment is strongly suggestive of a complex partial seizure. In the child not on antiepileptic medications, 3 minutes of hyperventilation usually precipitates an absence seizure. It is unusual for complex partial seizures to be precipitated by hyperventilation. Abnormalities documented on the EEG constitute the best confirmation of either seizure type.

Daydreaming is associated with boredom, can be “broken” with stimulation, and is not associated with motor activity. Absence seizures, however, also can be terminated with stimulation and tend to increase during periods of relaxation and tiredness. Tics and pseudoseizures may need to be considered, as well. Normal findings on an EEG study that includes several trials of 3–5 minutes of hyperventilation, however, virtually rule out absence seizures. Repeated studies or prolonged monitoring occasionally may be necessary when diagnostic confusion persists.

Etiology

Both acquired and inherited factors are implicated in the etiology of absence seizures, reflecting the heterogeneity of the patient population. Genetic factors predominate in children with typical absence seizures, whereas acquired disorders are more common in children with atypical absences. The genetics of absence seizures are discussed in Chapter 52.

Initial Evaluation

The extent of the diagnostic evaluation required in patients with absence seizures is variable and depends somewhat on findings from the history and physical examination. All patients need a neurologic examination, with hyperventilation, and an EEG. Patients who have normal development, normal findings on neurologic examination, a history suggestive of typical absence seizures, and 2.5- to 4-Hz spike-wave discharges on an otherwise normal-appearing EEG require no further studies. Children with developmental delay, an abnormal neurologic examination, a history suggestive of atypical absences, or an EEG showing slow spike-and-wave (less than 2.5 Hz) discharges, background slowing, or focal epileptiform discharges should have neuroimaging, preferably MRI, and possibly more specific tests, such as metabolic studies and a lumbar puncture.

Electroencephalographic Findings

The EEG signature of a typical absence seizure is the sudden onset of 3-Hz generalized symmetric spike-and-wave or multiple spike-wave complexes (Figure 53-2). The voltage of the discharges often is maximal in the frontal-central regions. The frequency tends to be faster, about 4 Hz, at the onset and may slow to 2 Hz toward the end of a discharge lasting longer than 10 seconds.

Hyperventilation is a potent activator of typical absence seizures [Adams and Lueders, 1981; Dalby, 1969]. Failure to induce an absence seizure with several trials of hyperventilation of 3–5 minutes’ duration in a child not receiving antiepileptic medication would make the diagnosis of typical absence seizures unlikely. Photic stimulation also may precipitate seizures, although the frequency of activation does not appear as high as with hyperventilation [Newmark and Penry, 1979].

When closely observed, children typically demonstrate behavioral changes with spike-and-wave discharges lasting longer than a few seconds. In studies of auditory reaction times, it has been found that responses are abnormal in 80 percent of the situations in which a stimulus is presented after onset of the spike-and-wave activity [Porter et al., 1973]. In a test of visual motor coordination in children with spike-and-wave activity on the EEG, discharges lasting longer than 3 seconds were associated with impaired function [Opp et al., 1992; Brown et al., 1974; Porter et al., 1973].

The interictal EEG in children with typical absence seizures typically shows a normal background. Brief bursts of generalized spike-and-wave activity, frontal spikes, or, less commonly, multifocal spikes can be seen. In absences, focal or multifocal spikes do not necessarily indicate a focal epileptiform process; rather, they may represent portions of generalized discharge that do not totally propagate uniformly to the cortex.

In atypical absences, the ictal EEG is more heterogeneous, showing 1.5- to 2.5-Hz slow spike-and-wave or multiple spike-and-wave discharges that may be irregular or asymmetric (Figure 53-3). Interictal EEG findings usually are abnormal, with background slowing and multifocal epileptiform features. The EEG features of typical and atypical absences are summarized in Table 53-3.

Pathophysiology

The observation that 3-Hz spike-and-wave discharges in absence seizures appear simultaneously and synchronously in all electrode locations led early investigators to speculate that the pathophysiologic mechanisms of absence seizures must involve “deep” structures with widespread connections between the two hemispheres.

A number of studies have demonstrated that the basic underlying mechanism in generalized absence epilepsies involves thalamocortical circuitry and the generation of abnormal oscillatory rhythms in this neuronal network [Coulter and Zhang, 1994; Crunelli and Leresche, 2002; Hosford et al., 1995; Huguenard and Prince, 1994b; Steriade et al., 1993; Steriade and Conteras, 1998]. The neuronal circuits that generate the oscillatory thalamocortical burst firing observed during absence seizures have now been identified. This circuit includes cortical pyramidal neurons, thalamic relay neurons, and the nucleus reticularis thalami (Figure 53-4) [Steriade and Llinas, 1988; Steriade et al., 1993]. The principal synaptic connections of the thalamocortical circuit include glutamatergic fibers between neocortical pyramidal cells and the nucleus reticularis thalami, gamma-aminobutyric acid (GABA)-ergic fibers from nucleus reticularis thalami neurons that activate GABA_A and GABA_B receptors on thalamic relay neurons, and recurrent collateral GABAergic fibers from nucleus reticularis thalami neurons that activate GABA_A receptors on adjacent nucleus reticularis thalami neurons. The nucleus reticularis thalami, therefore, is in a position to influence the flow of information between the thalamus and cerebral cortex [Snead, 1995]. The nucleus reticularis thalami cells undergo rhythmic burst firing during periods of sleep and continuous single-spike firing during wakefulness.

The cellular events that underlie the ability of nucleus reticularis thalami neurons to shift between an oscillatory and a firing mode include activation of the low-threshold, transient Ca²⁺ channels (T-channels) [Coulter et al., 1989, 1990; Crunelli and Leresche, 2002; Snead, 1995]. These channels appear to be important in the generation of absence seizures [Snead, 1995; Tsakiridou et al., 1995]. Mild depolarization of these neurons is sufficient to activate these T-channels and to allow the influx of extracellular Ca²⁺. Further depolarization produced by Ca²⁺ inflow often exceeds the threshold for firing a burst of action potentials. After T-channels are activated, they become inactivated rather quickly – hence the name transient. T-channels require a lengthy, intense hyperpolarization to reverse their inactivation (a process termed deinactivation). The requisite hyperpolarization can be provided by GABA_B receptors that are present on thalamic relay neurons. The interplay between GABA_B-mediated inhibition and the low-threshold T-type calcium channel therefore plays a critical role in generating the oscillating hyperpolarization/depolarization activity seen in the thalamus. In animal models of absences, GABA_B agonists produce an increase in seizure frequency (by facilitating deinactivation of T-channels), whereas GABA_B antagonists reduce seizure frequency.

Recurrent collateral GABAergic fibers from the nucleus reticularis thalami neurons activate GABA_A receptors on adjacent nucleus reticularis thalami neurons. Activating GABA_A receptors in the nucleus reticularis thalami, therefore, results in reduction of GABAergic output to the thalamic relay neurons and serves to reduce hyperpolarization and delay deinactivation of the T-channels. In animal studies, injection of the GABA_A agonists bilaterally into the nucleus reticularis thalami reduces absence seizure frequency. GABA_A activation of thalamic relay neurons, however, would be expected to have the opposite effect, increasing hyperpolarization and deinactivation of the T-channel.

Abnormal oscillatory rhythms could be caused by abnormalities of the T-channel or enhanced GABA_B function [Crunelli and Leresche, 2002; Snead, 1995]. In some animal models of absence seizures, T-channel activation in the nucleus reticularis thalami is significantly different than in control animals [Tsakiridou et al., 1995]. These aberrant T-channels may be one basis for absence seizures. In other animal models, an increase in GABA_B receptors in thalamic and neocortical neuronal populations, compared with that in controls, has been described [Hosford et al., 1995]. Because an increase in GABA in the thalamic relay neurons would lead to enhanced hyperpolarization and consequently deinactivation of the T-channels, GABA_B agonists produce an increase in seizure frequency, whereas GABA_B antagonists reduce seizure frequency in rodent models of absence epilepsy [Hosford et al., 1992; Liu et al., 1992; Marescaux et al., 1992; Tsakiridou et al., 1995].

Supporting these animal findings are the clinical observations that three drugs efficacious in the treatment of absence seizures – valproic acid, ethosuximide, and trimethadione– suppress T-currents [Coulter et al., 1990]. In addition, some clinical evidence indicates that vigabatrin, which increases endogenous GABA levels and thereby increases the activation of GABA_B receptors, worsens absence seizures in patients. Clonazepam, which activates GABA_A levels in the nucleus reticularis thalami, can be an effective anti-absence drug [Huguenard and Prince, 1994a].

Treatment

AED therapy is recommended for children with absence seizures. Although not life-threatening, absence seizures may lead to poor school performance, ridicule, and accidents. Because even brief generalized spike-and-wave discharges can affect cognitive function, it is therefore reasonable to begin drug therapy in most patients once the diagnosis is secure.

In a randomized, controlled clinical trial of 453 children with childhood absence epilepsy who were randomly assigned to one of these three drugs, ethosuximide and valproic acid were more efficacious than lamotrigine whereas there were no significant differences among the three drugs with regard to discontinuation because of adverse events (Glauser et al., 2010). However, ethosuximide had less effect on attention than valproic acid. This study would suggest that when considering both efficacy and side effects ethosuximide is the drug of first choice for childhood absence epilepsy. A single agent should be chosen and, after appropriate laboratory studies, initiated at a low dose and gradually increased. AED levels may be helpful, but dose changes in either direction should follow clinical indications. On dosage modifications, drug levels should be obtained only after sufficient time has elapsed to reach steady-state serum concentrations.

In the United States, many physicians will begin therapy with ethosuximide, primarily because of the rare, but severe, hepatotoxicity and pancreatitis associated with valproic acid [Dreifuss et al., 1987, 1989; Penry and Dean, 1993]. Ethosuximide has proved over the past two decades to be a safe and effective AED used almost exclusively for absence seizures [Schneider, 1993; Perucca, 1990]. Valproic acid is equally effective and generally is considered the drug of choice in the patient who has both absence and generalized tonic-clonic seizures, because ethosuximide is not effective in the treatment of generalized tonic-clonic seizures. Lamotrigine appears to be a promising drug in the treatment of absence seizures [Frank et al., 1999; French et al., 2004; Holmes et al., 2008].

Benzodiazepines, such as clonazepam, may be effective, but they usually are reserved for refractory cases because of the relatively high incidence of drowsiness and behavioral side effects [Pearl and Holmes, 2008]. Topiramate [Cross, 2002; Wheless, 2000], levetiracetam [Cohen, 2003], and acetazolamide [Reiss and Oles, 1996] may be useful. Carbamazepine [Yang et al., 2003], vigabatrin [Yang et al., 2003], and tiagabine [Knake et al., 1999] should be avoided because they may exacerbate absence seizures.

In addition to obtaining an interim history, the clinician should always have the child perform a hyperventilation test at follow-up visits. Eliciting a seizure during hyperventilation strongly indicates that the epilepsy disorder is not totally controlled [Adams and Lueders, 1981].

The duration of therapy is variable, although a general rule is to discontinue therapy gradually after 2 seizure-free years have elapsed. Continued presence of spike-and-wave activity on the EEG indicates a high risk for relapse following discontinuation of the drug. Normal EEG findings while the child is on therapy, however, cannot be used as a firm indication that the child will remain seizure-free, because the anti-absence drugs can suppress spike-and-wave activity on the EEG.

Prognosis

Approximately two-thirds of children with childhood absence epilepsy can be expected to enter long-term remission [Wirrell, 2003]. Sato et al. [1976] identified favorable prognostic signs in absence seizures as a negative family history of epilepsy, normal EEG background activity, and normal intelligence. Nearly 90 percent of children with these characteristics stopped having absence seizures.

Juvenile absence epilepsy may occasionally persist into adulthood, however, and juvenile myoclonic epilepsy, as discussed later on, does not spontaneously remit [Wirrell et al., 1996]. As a rule, onset of generalized tonic-clonic seizures before absence seizures carries a poorer prognosis than that noted with the reverse order [Sato et al., 1976].

Clinical Features

Clonic seizures are similar to generalized tonic-clonic seizures but are characterized by only rhythmic or semirhythmic contractions of a group of muscles. These jerks can involve any muscle group, although the arms, neck, and facial muscles are most commonly involved. The jerking often is asymmetric and irregular.

Differentiating a flurry of myoclonic seizures from a clonic seizure may be difficult. Clonic seizures, however, are associated with a more marked impairment of consciousness than is seen with a series of myoclonic seizures and typically include a postictal period, although this may be quite brief. Clonic seizures are more common in children than in adults and usually occur during the first few years of life.

Etiology and Prognosis

Clonic seizures are more likely to be symptomatic than are generalized tonic-clonic seizures. The prognosis is related to the etiology of the seizures.

Electroencephalographic Findings

The EEG may show fast activity, fast activity mixed with slow waves, or polyspike-and-wave discharges.

Clinical Features

Tonic seizures are brief seizures (usually lasting less than 60 seconds), consisting of the sudden onset of increased tone in the extensor muscles [Egli et al., 1985; Holmes, 1988]. If standing, the patient typically falls to the ground. Electromyographic activity is dramatically increased in tonic seizures.

Impairment of consciousness during the seizure is characteristic, although when seizures are brief, this change may be difficult to detect. Tonic seizures frequently are seen in patients with Lennox–Gastaut syndrome, a disorder consisting of a mixed seizure disorder, mental retardation, and the EEG findings of a slow spike-and-wave pattern [Aicardi, 1988a, b; Markand, 2003]. The seizures usually occur more frequently at night.

Tonic seizures frequently begin with a tonic contraction of the neck muscles, leading to fixation of the head in an erect position, widely opened eyes, and jaw clenching or mouth opening [Holmes, 1988]. Contraction of the respiratory and abdominal muscles often follows, sometimes leading to a high-pitched cry and brief periods of apnea. The tonic contractions may extend to the proximal musculature of the upper limbs, elevating the shoulders and abducting the arms. Asymmetric tonic seizures vary in severity, ranging from a slight rotation of the head to a tonic contraction of all of the musculature on one side of the body. Eyelid retraction, staring, mydriasis, and apnea also may occur [Aicardi, 1988b]. Occasionally, tonic seizures terminate with a brief clonic phase. Unlike in generalized tonic-clonic seizures, however, the clonic phase is abbreviated. Postictal impairment, with confusion, tiredness, and headache, is common. The degree of postictal impairment usually is related to the duration of the seizure.

Tonic seizures typically are activated by sleep and may occur repetitively throughout the night. They usually are more frequent during non-rapid eye movement (REM) sleep than during wakefulness and usually do not occur during REM sleep. Arousal from light sleep may occur after a tonic seizure. Because the child often does not wake up during the seizure, these seizures often go undetected.

Etiology and Prognosis

Tonic seizures usually occur in children with Lennox–Gastaut syndrome, which is discussed later.

Electroencephalographic Findings

The interictal EEG pattern in patients with tonic seizures usually is quite abnormal, consisting of slowing of the background, with multifocal spikes, sharp waves, and bursts of irregular spike-and-wave activity. The EEG ictal manifestations of tonic seizures usually consist of bilateral synchronous spikes of 10–25 Hz, of medium to high voltage, with a frontal accentuation. Simple flattening or desynchronization may also occur (Figure 53-5). Occasional multiple spike-and-wave or diffuse slow-wave activity may occur during a tonic seizure.

Clinical Features

Atonic (astatic) seizures, or “drop attacks,” are characterized by a sudden loss of muscle tone [Oguni et al., 1992, 1997]. They begin suddenly and without warning, and cause the patient, if standing, to fall quickly to the floor. Children with atonic seizures are more likely to fall backward than children with tonic seizures [Oguni et al., 1997]. Because muscle tone may be completely absent, the children have little means by which to protect themselves, and injuries often occur. The attack may be fragmentary and manifest as dropping of the head with slackening of the jaw or dropping of a limb.

At times, it may be difficult to distinguish epileptic from nonepileptic head drops. Children with head drops secondary to seizures often have a change in facial expression and subtle myoclonic extremity movements associated with the head drops. Head drops with a rapid head descent, followed by a slow recovery to the upright position, usually represent seizures [Brunquell et al., 1990].

In atonic seizures, a loss of electromyographic activity is characteristic. Consciousness is impaired during the fall, although the patient may regain alertness immediately on hitting the floor. Atonic attacks are frequently associated with myoclonic jerks either before, during, or after the atonic seizure [Egli et al., 1985; Schneider et al., 1970]. This combination has been described as myoclonic-astatic seizures. Atonic seizures are rare [Egli et al., 1985; Ikeno et al., 1985]; a majority of children with drop attacks have myoclonic or tonic seizures [Ikeno et al., 1985].

Etiology and Prognosis

Atonic seizures usually occur in children with Lennox–Gastaut syndrome or Doose’s syndrome. Doose’s syndrome is discussed in Chapter 56.

Electroencephalographic Findings

Atonic seizures usually are associated with rhythmic spike-and-wave complexes varying from slow, 1- to 2-Hz, to more rapid, irregular spikes or multiple spike-and-wave activity.

Electroencephalographic Findings

The hallmark of the EEG pattern in Lennox–Gastaut syndrome is the slow spike-and-wave discharge superimposed on an abnormal, slow background. The slow spike-and-wave or sharp-and-slow-wave complexes consist of generalized discharges occurring at a frequency of 1.5–2.5 Hz. The morphology, amplitude, and repetition rate may vary both between bursts and during paroxysmal bursts of spike-and-wave activity, and asymmetries of the discharge are frequent. The area of maximum voltage, although variable, usually is frontal or temporal in location. Often, sleep increases the number of epileptiform discharges, but these discharges may slow in frequency and become even more irregular than during the awake state (Figure 53-6). In REM sleep, the paroxysmal activity decreases markedly. Hyperventilation and photic stimulation rarely activate these discharges.

Etiology

The syndrome may arise de novo in a previously well child or may supervene in an already neurologically or medically handicapped patient. Lennox–Gastaut syndrome has been divided into primary and secondary cases. Primary refers to cases in which the etiology is idiopathic, whereas secondary refers to cases in which the disorder is symptomatic of a definable etiology. Markand [1977] was able to identify an underlying disorder in 64 percent of 83 patients, and Chevrie and Aicardi [1972] were able to do so in 66 percent of 80 patients. Other investigators, however, have reported a smaller incidence of identifiable causes in Lennox–Gastaut syndrome [Kurokowa et al., 1980; Schneider et al., 1970]. The causes of Lennox–Gastaut syndrome are listed in Box 53-1 and are similar to those seen in infantile spasms. In fact, in many patients with infantile spasms, (30–40 percent) the disorder will progress to Lennox–Gastaut syndrome [Kurokawa et al., 1980; Olmos-Garcia de Alba et al., 1984; Riikonen, 1996]. In some children with infantile spasms that evolve into Lennox–Gastaut syndrome the epileptic spasms continue to occur in series and become one of the seizure types (Donat and Wright, 1991).

Although a majority of patients with Lennox–Gastaut syndrome have a static disorder, degenerative disorders may occasionally manifest as this syndrome [Markand, 1977; Pampiglione and Harden, 1977]. Genetic factors do not play a significant role in a majority of individuals with this disorder.

Treatment

Management of the child with Lennox–Gastaut syndrome is difficult. Because the degree of heterogeneity with this syndrome is considerable, the response to AED therapy is variable. Individualized treatment approaches are necessary. Usually no single AED is effective in controlling all of the seizure types in Lennox–Gastaut syndrome. This frequently leads to polypharmacy. In addition, the sedating side effects of the AEDs may lead to an increase in seizure frequency. The other notable problem is tolerance. In patients who have an initial improvement, loss of efficacy often occurs after a few months of treatment. This decreased efficacy usually leads to an increase in dosage, often with accompanying side effects, or to additional polytherapy. Although polytherapy sometimes cannot be avoided, the physician must be aware of the potential for harm. Withdrawal of AEDs, particularly the barbiturates and benzodiazepines, may lead to improved alertness and an actual reduction in seizure frequency.

In the United States, valproate likely is the most commonly used drug in the treatment of Lennox–Gastaut syndrome. It is a broad-spectrum drug with effectiveness against all of the common seizure types in the syndrome. The lack of sedative side effects and the broad spectrum of action make the drug appealing. The drug has been associated with hepatotoxicity, however, particularly in children younger than 2 years of age. Phenytoin is considered by some investigators to be the drug of choice for tonic seizures. Intravenous phenytoin is used as a first-line drug for status epilepticus in Lennox–Gastaut syndrome (Richard et al., 1993) since benzodiazepines can precipitate tonic status epilepticus in Lennox–Gastaut syndrome [Tassinari et al., 1972].

Felbamate, released in the United States in 1993, was the first drug to be demonstrated to be effective in Lennox–Gastaut syndrome [Dodson, 1993a, b; The Felbamate Study Group in Lennox–Gastaut Syndrome, 1993]. After the drug was released, it was found to result in fatal aplastic anemia and hepatic failure. The overall risk of aplastic anemia for patients taking felbamate is estimated to be 1:5000, 100 times greater than the risk in the normal population, and the risk of fatal outcome is approximately 1:10,000 [Kaufman et al., 1997]. Felbamate-associated aplastic anemia has not been reported in children younger than 13 years [Brodie and Pellock, 1995; Pellock and Brodie, 1997]. Risk factors associated with aplastic anemia include a history of blood dyscrasia or of substantial toxicity to other AEDs, serologic or clinical evidence of an autoimmune disorder, and a history of treatment with felbamate for less than 1 year [Pellock, 1999].

Both lamotrigine [Donaldson et al., 1997; Motte et al., 1997] and topiramate [Guerreiro et al., 1997; Ormrod and McClellan, 2001; Sachdeo et al., 1999] have demonstrated efficacy in Lennox–Gastaut syndrome but rarely with total control of the seizures.

In a double-blind, placebo-controlled trial of lamotrigine in 169 patients with Lennox–Gastaut syndrome, either lamotrigine or placebo was added to the maintenance medications. After 16 weeks, 33 percent of children receiving lamotrigine and 16 percent of placebo recipients had more than a 50 percent reduction in seizure frequency [Motte et al., 1997]. In a double-blind, placebo-controlled, multicenter trial of topiramate for adjunctive therapy for Lennox–Gastaut syndrome, the frequency of drop attacks was reduced compared with that in the placebo group [Sachdeo et al., 1999].

The ketogenic diet (see Chapter 60), one of the oldest methods of treating childhood epilepsy, remains a reasonable therapy for children with seizures refractory to standard drug therapy [Vining et al., 1998; Wheless, 2004; Ferrie and Patel, 2009]. Although the diet was considered a therapy of last resort for many years, recent resurgence in interest has occurred, primarily owing to its success in several highly publicized cases. The diet consists of a high proportion of fats and small amounts of carbohydrate and protein, with a fat to carbohydrate and protein ratio of 4:1. Although it is clear that the child must remain in a state of ketosis for the diet to be effective, the basis of the therapeutic effectiveness of the ketogenic diet remains uncertain.

The resurgence in interest in the diet is appropriate because it does provide an attractive alternative in the treatment of childhood epilepsy. The literature supports the view that the ketogenic diet improves seizure control in a significant number of children with medically intractable epilepsy [Prasad et al., 1996; Rubenstein and Vining, 2004]. One-third to one-half of children appear to have an excellent response to the ketogenic diet in terms of a marked or complete cessation of seizures or reduction in seizure severity. In another one-third of the children, a partial reduction in seizure frequency or severity occurs; the remaining children have no appreciable benefit from the diet. Improvement in alertness and behavior often is seen when the child is placed on the diet. It is not clear whether this benefit is secondary to withdrawal of drugs, reduction in seizure frequency, or a direct result of the diet.

Aspects of the surgical evaluation of children with generalized epilepsies, including use of vagal nerve stimulators, are reviewed in Chapter 61.

References

The complete list of references for this chapter is available online at www.expertconsult.com.

See inside cover for registration details.