As discussed in Chapter 1, classification of epileptic seizure types has practical importance but also significant limitations. Historically used terms may not accurately reflect the complex pathophysiology or phenomenology of seizures. Generalized seizures involve synchronous epileptic discharges throughout the cerebral cortex, and are further distinguished by the terms primary or secondarily. Seizures that appear to have synchronous bihemispheric involvement at initiation (based on clinical and electrographic evidence) are primary generalized seizures and those that begin in a focal area of cortex and subsequently spread to involve both hemispheres are secondarily generalized. This distinction is not always straightforward, and a high index of suspicion should be maintained for focal onset seizures that appear generalized, because evaluation and treatment differs between generalized and focal epilepsies.

By age 2–4 years, children can manifest all of the generalized seizure types that are seen in older age groups—generalized tonic–clonic seizures (GTCs), absence seizures, myoclonic seizures, tonic seizures, and atonic/astatic seizures. Infantile spasms do not typically present at this age, and the occasional cases of late-onset infantile spasms are reviewed in Chapter 28.

Generalized seizures presenting in early childhood often fail to meet criteria for an epilepsy syndrome, or the syndrome may not be obvious at presentation.

The process of evaluation for a child in this age group, as with any child with epilepsy, begins with an accurate diagnosis of seizure, seizure classification, and, when possible, determination of etiology. A series of questions (see Box 11–1) should lead to the most appropriate diagnosis, which then guides further diagnostic evaluation and treatment decisions. This chapter is organized by clinical seizure type, and will focus on young children presenting with two or more unprovoked generalized seizures who do not fit into a definitive epilepsy syndrome.

CLASSIFICATION AND DIAGNOSIS

History

Most GTCs in this age group are, in fact, something else (FIG. 11-1). The first step in approaching the young child presenting with an event reported to be a GTC is to obtain a detailed description of the episode. Witnesses may describe a wide variety of clinical events (epileptic and nonepileptic) as “grand mal seizures” or “convulsions.” In early childhood, events mimicking GTCs include breath-holding spells, syncope with subsequent convulsive movements, and gastroesophageal reflux. A careful description is essential for distinguishing epileptic from nonepileptic events. For example, breath-holding spells may be provoked by crying or a sudden fright, and are characterized by a respiratory pause in expiration resulting in loss of consciousness or limpness, sometimes followed by opisthotonic posturing or clonic limb movements.¹

A common cause of a GTC in a 2–4-year old is a focal onset seizure with rapid secondary generalization, so the diagnostic evaluation should carefully evaluate this possibility. Distinguishing primary from secondarily generalized seizures based on the history of the event is often challenging. Clues to focal onset may be missed because young children are often unable to articulate an aura, and parents may be so overwhelmed by seeing the convulsion that they are unable to describe details of seizure onset or postictal findings such as weakness. Requesting a demonstration of what the parent or other witness saw may reveal clues not apparent in a verbal description of the events.

The next step in seizure classification is taking a careful developmental and medical history, focusing on risk factors for epilepsy such as premature birth, neonatal hypoxic–ischemic encephalopathy, neonatal seizures, significant head trauma, CNS infection, stroke, or developmental delays. While these factors do not exclude the possibility of a primary GTC, their presence should raise suspicion for symptomatic focal or multifocal epilepsy characterized by rapidly secondarily generalized seizures. Similarly, the neurologic exam should be directed toward identifying focal neurologic deficits suggesting focal epilepsy. Though handedness is established by age 36 months in most children, significant functional asymmetries and developmental reflex asymmetries (e.g., the parachute reflex) should raise red flags about the possibility of an underlying focal lesion.

When clues to seizure classification are not forthcoming from a description of the event, medical history, and neurologic exam, further diagnostic studies are needed.

Electroencephalography (EEG)

EEG is a helpful tool for refining basic seizure classification. Every child with an unprovoked GTC should have a routine EEG searching for background abnormalities, focal epileptiform features, or generalized epileptiform abnormalities. The recording should include photic stimulation, and, in cooperative older children, hyperventilation. In our experience, it is possible to coax a child as young as 3 years into blowing on a pinwheel for a short period of time, and children 4 years and older are able to cooperate more reliably with hyperventilation. A focal EEG abnormality strongly implies that a presumed GTC was a focal seizure with secondary generalization. A normal EEG, particularly if sleep and arousal from sleep were recorded, may also suggest a focal onset seizure. The presence of generalized epileptiform abnormalities such as spike-wave discharges, polyspike-wave discharges, or generalized paroxysmal fast activity, suggests a true primary generalized seizure.

If basic seizure classification remains unclear and response to initial antiepileptic drug (AED) treatment is suboptimal, then ictal recordings in an epilepsy monitoring unit may be needed. If seizures are too infrequent for ictal recording on an inpatient unit to be practical, a longer recording in the monitoring unit may still yield interictal clues not seen on routine EEG. While ictal recording with an ambulatory EEG may be attempted, it is often challenging to obtain high-quality recordings using ambulatory EEG in this age group.

Differential Diagnosis

Most idiopathic generalized epilepsy (IGE) syndromes that are characterized by GTCs do not present between 2 and 4 years of age. Identifying a recognized syndrome often depends on characterizing the other seizure types in a child with GTCs. The most likely syndrome diagnosis for a child presenting with a GTC at this age may be myoclonic–astatic epilepsy (MAE; Doose syndrome), which often begins with a febrile or afebrile GTC followed by typical myoclonic or myoclonic–astatic seizures days to months later.² Thus, the syndrome may not be initially obvious.

The syndrome of primary GTCs upon awakening (PGTCA) typically presents after age 6 years.³ Juvenile absence epilepsy (JAE) and juvenile myoclonic epilepsy (JME), syndromes in which children may have GTCs, also present later in life. Typical childhood absence epilepsy (CAE) rarely presents under age 3 or 4, and does not involve early GTCs.⁴ IGE with absences of early childhood, an epilepsy syndrome distinct from CAE, may involve early GTCs in two-thirds of patients. Some children with a form of “borderline severe myoclonic epilepsy of infancy (SMEB)” may have only GTCs—this rare epilepsy type, labeled “severe idiopathic generalized epilepsy of infancy with GTCs” presents before age 2 years.⁵ GTCs are not the predominant seizure type in Lennox–Gastaut syndrome (LGS).

After thorough evaluation, many children with GTCs between 2 and 4 years of age will not meet criteria for any recognized epilepsy syndrome and will not have a clear symptomatic etiology. These children can be described as having “generalized epilepsy characterized by GTCs, etiology unknown.” This diagnosis is imprecise, but recognizes the many uncertainties surrounding the clear diagnosis of GTCs in this age group.

The search for an etiology of GTCs also begins with a thorough history, and general physical and neurologic examination. The same points of the history and exam that aid in seizure classification also provide information for assisting the search for an underlying cause.

Imaging Studies

Brain imaging is recommended for children presenting with a GTCs if a focal abnormality is found on neurologic exam or EEG. In the developmentally normal child with a normal neurologic exam, and EEG characteristic of an IGE, imaging may not be required.⁶ Children presenting to an emergency room with a first GTC are often imaged with computed tomography (CT) scan to rule out hemorrhage, mass lesions, or hydrocephalus, although such studies are low yield except in certain high-risk groups, such as children with general medical problems or closed head injury.⁷ In the nonemergent setting, when imaging is indicated, an MRI is the modality of choice. With a high-quality MRI, some children with GTCs and generalized epileptiform EEG patterns may actually have evidence of unilateral or bilateral structural lesions, often malformations of cortical development (MCDs). It is unclear why some children with focal or multifocal structural lesions have apparently generalized epilepsy on clinical and EEG evaluation. Possibilities include an age-related susceptibility to rapid secondary synchrony of focal epileptic activity, or the limitation of current diagnostic techniques to distinguish among focal, multifocal, and generalized epilepsy.

Other Tests

Beyond EEG and MRI, other diagnostic studies are generally not required in a normal toddler or preschool age child with new-onset, self-limited unprovoked GTCs. In children with abnormal examinations or developmental delays, the clinical presentation will dictate which laboratory tests are needed to evaluate for genetic or metabolic disorders. The role of epilepsy-specific genetic testing in a normal child presenting with GTCs is presently unknown. The spectrum of clinical manifestations of SCN1A mutations in childhood epilepsy is increasingly wide, extending beyond the already broad boundaries of generalized epilepsy with febrile seizures plus (GEFS+), and Dravet syndrome.⁸ The potential clinical utility of SCN1A testing for probands, parents, and siblings must be carefully considered if such testing is to be performed on young children with isolated GTCs.⁹

If a diagnosis of “generalized epilepsy—characterized by GTCs” is established, then treatment with antiepileptic drugs (AEDs) taken daily is usually indicated (Table 11–1). There is little data specific to GTCs in young children. In general, valproic acid (VPA), lamotrigine (LTG), topiramate (TPM), and levetiracetam (LEV) are preferred, due to their broad spectrum of efficacy against multiple seizure types. Zonisamide (ZNS), clobazam (CLB), rufinamide (RFM), and phenobarbital are additional options. Carbamazepine (CBZ), oxcarbazepine (OXC), and gabapentin (GBP) are less useful, because they may precipitate other generalized seizure types, such as absence or myoclonic seizures.¹⁰

The goal of AED treatment in children with GTCs is to prevent seizure recurrence while minimizing adverse effects. Since there is no data comparing the efficacy of AEDs for the treatment of GTCs in early childhood, decisions regarding specific AEDs are based on available evidence for efficacy, safety, tolerability, titration schedule, ease of administration, and cost. Because knowledge about the effectiveness of individual AEDs on specific seizure types and patient populations changes rapidly, we advise that the information here be used only as a guide to current practice.

LEV is a commonly used AED in children with new-onset seizures of many types, including GTCs in primary generalized epilepsies and focal epilepsies. It is often preferred over other AEDs for its broad range of efficacy, its favorable tolerability profile, its ability to be titrated quickly, and its lack of drug–drug interactions. LEV has US FDA indications for use in primary GTCs in children as young as 4 years of age,¹¹ and its efficacy in children under the age of 2 with refractory partial or generalized seizures has also been demonstrated.¹² The use of LEV as initial monotherapy in children has been described, but largely in retrospective studies that include generalized and focal seizure types, with very few patients in the 2–4-year-age range.^13,14 Nevertheless, based on extrapolation to this population from available data, and because of its excellent tolerability profile, LEV is often preferred as the first-line drug for new-onset GTCs in children. Its efficacy against a broad range of seizure types is particularly appealing because, as discussed earlier, the primary versus secondary nature of GTCs in this age group is often unclear. The side effect most often reported is behavioral disturbances or irritability, which can occur in as many as 30% of children under age 4, but leads to discontinuation of the medication in only 16%.¹⁵

LTG also has efficacy for a broad range of seizure types including GTCs, and has been shown to be efficacious in treating primary generalized seizures in children as young as age 2 years.¹⁶ LTG is generally well tolerated, but titration to a therapeutic dose must be slow to minimize the risk of a serious rash. Serious rashes due to LTG are rare, particularly with the recommended titration schedules, but children have a three-fold greater risk than adults.¹⁷ The requirement for slow titration may limit the use of LTG in children with frequent GTCs, unless another medication such as a long-acting benzodiazepine is used as a bridge during the initiation period.

TPM is also efficacious in treating newly diagnosed epilepsy, including epilepsies characterized by GTCs in children as young as 2 years.¹⁸ Most side effects are dose dependent, and include weight loss, difficulties with attention or concentration, sleepiness, and paresthesia. Rare idiosyncratic side effects include glaucoma and urolithiasis. An issue of practical importance in using TPM in this age group is the rapid rate of clearance,¹⁹ which can necessitate using dosing ranges far higher than in older age groups (as high as 20 mg/kg per day in some toddlers in our experience), and three times daily dosing schedules.

VPA until recently was often used as the first-line treatment for GTCs, particularly when a primary generalized epilepsy was suspected or could not be ruled out. Though decades of experience with this agent have established it as a safe medication in most populations, certain idiosyncratic and dose-related side effects merit comment. The most serious is irreversible hepatotoxicity, an idiosyncratic reaction with an incidence of less than 1:16,000 in children aged 3–10 years on monotherapy, and less than 1:8000 in children over age 2 years on multiple medications.²⁰ Infants under the age of 2 years on multiple AEDs are at greatest risk, with a 1:600 incidence of hepatotoxicity. Underlying metabolic disease is believed to contribute to the risk of hepatotoxicity, and at least one specific mitochondrial disorder is known to be highly associated with valproate-related liver toxicity.²¹ Routine laboratory surveillance can help identify dose-related toxicities such as thrombocytopenia, but is unlikely to identify idiosyncratic reactions like liver failure if the patient is not symptomatic. Close clinical follow-up and parental education about red-flag symptoms are more important than frequent laboratory testing for identifying idiosyncratic hepatic dysfunction in high-risk patients. No studies have clearly defined a role for the preventative use of levocarnitine in children treated with valproate, but some recommend routine oral supplementation, particularly those with the highest risk category.²²