Seizures and epilepsy in children: Initial treatment and monitoring

INTRODUCTION — Children with epilepsy, particularly infants, differ from adults not only in the clinical manifestations of their seizures, but also in the presence of unique electroencephalogram (EEG) patterns, etiologies, and response to antiseizure medications. The immature brain, particularly in the neonate and young infant, differs from the adult brain in the basic mechanisms of epileptogenesis and propagation of seizures. It is more prone to seizures, but seizures are more apt to disappear as the child grows.

This topic presents an overview of the initial treatment of seizures and epileptic syndromes in children. Other aspects of seizures and epilepsy in children are presented separately:

Seizures and epilepsy in children: Classification, etiology, and clinical features
Seizures and epilepsy in children: Clinical and laboratory diagnosis
Epilepsy syndromes in children
Self-limited focal epilepsies of childhood
Seizures and epilepsy in children: Refractory seizures
Epilepsy in children: Comorbidities, complications, and outcomes

Neonatal seizures and epilepsy are reviewed elsewhere. (See "Overview of neonatal epilepsy syndromes" and "Clinical features, evaluation, and diagnosis of neonatal seizures" and "Treatment of neonatal seizures" and "Etiology and prognosis of neonatal seizures".)

WHEN TO START ANTISEIZURE MEDICATION THERAPY — A child's first seizure may be caused by an acute illness, such as a metabolic derangement or infectious disorder, and be nonrecurrent, or may represent the beginning of epilepsy. A decision must be made about initiating chronic antiseizure medication treatment if a potentially reversible acute cause is not found during the evaluation. (See "Seizures and epilepsy in children: Clinical and laboratory diagnosis" and "Seizures and epilepsy in children: Clinical and laboratory diagnosis", section on 'Setting in which episodes occur'.)

Unprovoked seizure — The term "unprovoked seizure" refers to a seizure of unknown etiology as well as one that occurs in relation to a preexisting brain lesion or progressive nervous system disorder (often referred to as a remote symptomatic seizure). Unprovoked seizures are distinct from seizures due to an acute condition such as a toxic or metabolic disturbance, fever, head trauma, or acute stroke (ie, acute symptomatic seizures). (See "Seizures and epilepsy in children: Classification, etiology, and clinical features", section on 'Definitions'.)

The approach to starting antiseizure medication therapy for a first-ever versus a second unprovoked seizure is reviewed in the sections that follow.

First unprovoked seizure

●In most cases, we refrain from starting antiseizure medication therapy for a child with a first unprovoked seizure. This approach is consistent with the current practice of most neurologists. Similarly, a consensus statement from the International League Against Epilepsy (ILAE) states that in an otherwise well infant, a policy of "wait and see" with close follow-up monitoring is reasonable after a first afebrile seizure [1]. Even if the child has a static encephalopathy, treatment can be withheld until a recurrence pattern is established.

●Antiseizure medication treatment may be warranted after a first unprovoked seizure for select children with specific epilepsy syndromes [2]. As an example, treatment with a broad spectrum antiseizure medication would be indicated for a teenager with a first-time generalized tonic-clonic seizure whose EEG showed 4 to 6 Hz generalized spike and slow wave activity, confirming the diagnosis of juvenile myoclonic epilepsy (JME). In addition, the risk of seizure recurrence is higher in children with a potential remote symptomatic etiology, particularly if the seizure was focal and the electroencephalogram (EEG) or brain magnetic resonance imaging (MRI) is abnormal; in such cases, the benefits of immediate antiseizure medication therapy may outweigh the risks.

In all cases, the decision should be individualized, weighing the risks of recurrent seizure against the potential risks and benefits of antiseizure medication therapy and incorporating patient values and preferences. Treatment is reasonable when the benefits of reducing the risk of a second seizure are greater than the risks of pharmacologic adverse effects [3].

Factors influencing the decision — The main factors to consider in deciding whether or not to treat a child with a first-time unprovoked seizure include:

•The risk for recurrent seizures, which varies based on clinical factors. (See 'Risk of seizure recurrence' below.)

•The relative risk reduction that can be expected from immediate antiseizure medication therapy. Limited data in children suggest that early versus delayed antiseizure medication therapy reduces the short-term risk of a recurrent seizure but does not appear to affect the long-term prognosis of epilepsy. Specifically, starting antiseizure medication treatment after a first unprovoked seizure does not confer additional benefit for long-term seizure control compared with starting after a second seizure [3]. (See 'Effects of early versus deferred therapy' below.)

•The risks of not treating, which include another seizure with its attendant risk of injury and psychologic stigma (including restrictions on driving and work environment in teenagers and young adults) and, infrequently, status epilepticus.

•The risks of chronic antiseizure medication therapy, which includes possible effects on school performance and behavior, allergic reactions, and systemic toxicity. The financial burden of chronic antiseizure medications, office visits, and laboratory tests should also be considered.

Risk of seizure recurrence — The child who is neurologically normal, has no history of a prior neurologic illness, and has an unprovoked seizure with no evident acute cause has an approximately 25 percent risk of having another seizure in the next year and a 45 to 50 percent risk over the next three years [4-7].

Clinical factors associated with an increased risk of recurrent seizures include:

●Prior neurologic insult (ie, remote symptomatic seizure)

●Focal seizure

●Significant brain MRI findings

●Abnormal EEG

●Family history of epilepsy

The magnitude of risk associated with each one of these factors varies, and the additive effects of multiple risk factors have not been clearly established. In a large prospective study of 283 children with a first-time unprovoked seizure, neurologically normal children with no history of prior neurologic illness had a 24 percent risk of having a seizure in the next year [4,5]. The one-year recurrence risk increased to 37 percent in children with a prior neurologic insult (remote symptomatic), such as cerebral palsy, and increased to 70 percent in patients who had two seizures separated by at least 24 hours. Most but not all of the children in this study were not treated with an antiseizure medication.

In the same study, the one-year recurrence risk was 41 percent if the EEG was abnormal (epileptiform activity, focal or generalized slowing) compared with 15 percent in children with a normal EEG [4].

In another study, significant MRI findings, present in 16 percent of children, were associated with a twofold higher incidence of recurrent seizure by 27 months (80 versus 44 percent) and were more predictive of short-term seizure recurrence than EEG findings [8].

Status epilepticus as the presenting seizure may be another risk factor for recurrent seizures, although the available data are less consistent [4,9].

Effects of early versus deferred therapy — In adults, immediate antiseizure medication therapy after a first-time unprovoked seizure reduces the risk of recurrence by approximately 35 percent at one to two years postseizure [10]; more limited data in children suggest that the short-term benefits of antiseizure medication therapy are similar to those in adults.

Withholding treatment until after the second seizure does not alter the long-term prognosis of epilepsy, however. A study that randomly assigned 419 patients with a first tonic-clonic seizure to immediate antiseizure medication therapy or treatment only after a second seizure found that immediate treatment reduced the short-term relapse rate [11]. However, at one and two years, the number of patients who remained seizure free in the early-treatment and delayed-treatment groups were similar (83 to 87 percent at one year, 60 to 68 percent at two years). A similar study of 1443 patients (38 percent were ≤19 years old) found that immediate antiseizure medication treatment reduced the short-term (one- to two-year) risk of seizure relapse, but had no effect on the long-term risk of relapse [12]. The benefit of immediate treatment was lost by four years after a first seizure and by six years after multiple seizures.

The long-term risk of mortality after a single seizure appears to be quite low, and there is no evidence that treatment after an initial seizure has any impact on mortality. In a cohort study of 407 children with a first unprovoked seizure, four had a seizure-related death over a 14-year observation period [5]. Each of these had multiple seizures that were not fully controlled with antiseizure medication treatment; none died prior to initiation of antiseizure medication therapy, and two had been treated after the first seizure that was status epilepticus.

Second unprovoked seizure — We start antiseizure medication therapy for most children who present with a second unprovoked seizure, since seizure recurrence indicates that the patient has a substantially increased risk for additional seizures (ie, epilepsy) [13].

There are some exceptions, however. Many parents and caregivers elect no antiseizure medications if the seizures are infrequent and/or mild. However, the definitions of "infrequent" and "mild" are subjective and may vary among parent and caregivers. By contrast, children with absence seizures, atonic seizures or drop attacks, and infantile spasms are virtually always treated, since they usually present to the clinician with an already established pattern of frequent seizures.

Acute symptomatic seizure — An acute symptomatic seizure (also referred to as a provoked seizure or reactive seizure) is a seizure that occurs in close temporal association with an acute systemic illness or brain insult. Children who have a seizure in the setting of an acute illness (eg, acute infection, acute head injury) have a low risk of seizure recurrence compared with other children with a first seizure [14]. When a seizure is associated with a specific underlying acute etiology, seizure recurrence is likely only if the underlying etiology recurs. Examples include seizures associated with febrile illnesses, metabolic derangements (eg, hyponatremia), and concussion.

Management of acute symptomatic seizures should be focused on correction of the acute provoking illness or anomaly and preventing its recurrence. Children with an acute symptomatic seizure may not require antiseizure medication therapy. Exceptions include hospitalized children who are recovering from an acute illness that triggered seizures (eg, correctable conditions such as severe electrolyte disturbances with dehydration, arginine vasopressin deficiency (AVP-D, previously called central diabetes insipidus), syndrome of inappropriate secretion of antidiuretic hormone secretion [SIADH]); they may be loaded and maintained on an antiseizure medication for several days to a week or two, but antiseizure therapy is usually stopped before hospital discharge. Children with acute seizures due to meningitis or encephalitis are more likely to stay on antiseizure medication therapy for a little longer after hospital discharge. Children with acute posttraumatic seizures after a concussion often do not require any antiseizure therapy. Children with severe traumatic brain injury requiring intensive care will usually receive maintenance antiseizure medication coverage regardless of whether they have had seizures.

Febrile seizure — Children presenting with a first-time febrile seizure have an approximately 30 to 35 percent chance of having another febrile seizure during early childhood. Although antiseizure medications have been shown to lower the risk of recurrent febrile seizures, given the benign nature of febrile seizures, the risks of side effects from antiseizure medications generally outweigh the benefits for most patients. This is discussed in more detail separately. (See "Treatment and prognosis of febrile seizures", section on 'Role of preventive therapy'.)

SELECTION OF AN ANTISEIZURE MEDICATION — There is an ever-growing list of antiseizure medications and nonpharmacologic therapies available to manage childhood epilepsy [15]. Traditionally, the medications have been separated into "older" and "newer" groups based upon their historic regulatory approval and availability. Typically, when a medication is first approved for epilepsy, it receives an "on-label indication" for add-on (adjunctive) therapy for partial-onset seizures in adults. Then, as experience grows and other studies are done, the use of the drug may expand to other seizure types and younger age groups.

Seizure-related considerations — The antiseizure medication chosen for initial therapy should be one that is effective for a particular seizure type or syndrome [16,17]. The initial work-up should include an attempt to determine the specific type of seizure the child experienced and, if possible, the epilepsy syndrome. The optimal choice of antiseizure medication depends on the type of seizure and the epilepsy syndrome. (See "Seizures and epilepsy in children: Classification, etiology, and clinical features" and "Epilepsy syndromes in children".)

For some epilepsy syndromes, available data support choosing certain antiseizure medications over others as first-line therapy. Examples include:

●Corticotropin (ACTH) for infantile spasms (see "Infantile epileptic spasms syndrome: Management and prognosis", section on 'Corticotropin (ACTH)')

●Ethosuximide or valproate for childhood absence epilepsy (see "Childhood absence epilepsy", section on 'Treatment')

●Valproate for juvenile myoclonic epilepsy (see "Juvenile myoclonic epilepsy", section on 'Valproate')

For other types of epilepsy, such as focal epilepsy due to a remote symptomatic cause or mesial temporal sclerosis, there are no clear differences in efficacy among various antiseizure medications, and clinicians generally choose first-line therapy based on other drug-related factors such as side effects, cost, and dosing intervals. Levetiracetam, which has broad-spectrum activity and has minimal drug interactions, has become a common choice for first-line treatment of early-life epilepsy, although other drugs are equally reasonable and the choice should be individualized [18]. (See 'Drug-related considerations' below.)

In some cases, particularly for generalized epilepsy syndromes, seizures may be aggravated by the administration of a narrow-spectrum antiseizure medication, when a broad-spectrum antiseizure medication is more appropriate (table 1). As examples, carbamazepine and phenytoin have been reported to worsen absence and myoclonic seizures in individuals with idiopathic generalized epilepsy [19-22]. (See "Epilepsy syndromes in children" and "Childhood absence epilepsy", section on 'Drugs to avoid' and "Juvenile myoclonic epilepsy", section on 'Antiseizure medications to avoid'.)

Drug-related considerations — Drug-related factors include dose formulation, dose frequency, the relative risk of certain side effects, and the potential for drug-drug interactions. Single-drug therapy is the goal of epilepsy treatment. Monotherapy is associated with better compliance, fewer adverse effects, less potential for teratogenicity, and lower cost than is polytherapy. Drug interactions are avoided and pharmacokinetics are simplified.

In the absence of clear differences in efficacy among various antiseizure medications, clinicians must choose first-line therapy primarily based on pharmacokinetics, adverse effects, and consideration of drug-drug interactions. Cost effectiveness is also desirable; as an example, the World Health Organization recommends phenobarbital as the treatment of choice for partial and tonic-clonic seizures in countries with restricted resources [23]. The pharmacology and specific indications for individual antiseizure medications are discussed separately. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects".)

Pharmacokinetics and formulation

●Half-life – A long serum half-life allows for relatively smooth serum levels with less frequent daily dosing that can enhance medical compliance (table 2A-B). Examples of drugs that require only once- or twice-daily dosing include antiseizure medications that come in a sustained-release form (eg, phenytoin, carbamazepine, valproic acid, levetiracetam, lamotrigine) or have particularly long half-lives (ethosuximide, phenytoin, phenobarbital, zonisamide). These antiseizure medications allow for "make-up dosing." As an example, if a child on ethosuximide misses the once-a-day dose, it can be taken the next day with the regularly scheduled dose.

●Elimination kinetics – Elimination kinetics also should be considered when choosing an antiseizure medication. With linear or first-order kinetics, a constant fractional amount of the drug is eliminated over time, independent of the concentration of the drug in the serum. With nonlinear, dose-dependent, saturable, or concentration-dependent kinetics, as the level of the drug increases, clearance declines as the elimination mechanisms become saturated. Linear kinetics is more desirable.

Valproate, carbamazepine, and phenytoin have nonlinear kinetics. Phenytoin is particularly troublesome, with linear kinetics at low serum levels, and nonlinear kinetics as the serum level approaches the low- to mid-therapeutic range. Thus, a small increase in dose may lead to a large and potentially toxic increase in the serum level. At high serum levels, phenytoin's half-life is significantly longer. At these higher levels, it may take only a minimal increase in the daily dose to attain the target level; in some cases, this increase is made every other day.

●Formulations – In infants and younger children, oral suspensions, chewable tablets, and sprinkle formulations may be useful.

Adverse effects — The adverse effects of antiseizure medications make a significant contribution to reduced quality of life in patients with epilepsy. While many antiseizure medication adverse effects seem to be common to this entire class of medicines (eg, drowsiness, dizziness, diplopia, and imbalance), others are more specific to an individual drug. These should be considered in selecting an antiseizure medication, since certain adverse effects are either more likely or more problematic in certain patients.

●Common neurotoxic and systemic side effects are summarized in the table (table 3).

●Less common, often idiosyncratic, but potentially serious adverse events are summarized separately (table 4).

The adverse effects of individual drugs are discussed in detail separately. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects".)

Teratogenicity — There is an increased risk of major and minor congenital malformations in fetuses exposed to antiseizure medications. These risks are best characterized for valproate, phenobarbital, phenytoin, and topiramate, but other antiseizure medications may also be implicated, particularly in the context of antiseizure medication polytherapy. In addition, some antiseizure medications have been associated with long-term neurocognitive effects and an increased risk for autism spectrum disorders. These and other risks are discussed in detail separately. (See "Risks associated with epilepsy during pregnancy and the postpartum period", section on 'Effects of ASMs on the fetus and child'.)

Growing awareness of the fetal risks of in utero exposure to valproate in particular, together with recognition that valproate is one of the most effective drugs for certain types of epilepsy that commonly affect adolescent girls (eg, juvenile myoclonic epilepsy), have led a joint task force of the Commission on European Affairs of the International League Against Epilepsy (ILAE) and the European Academy of Neurology to issue a special report to provide guidance on the use of valproate in girls and women of childbearing potential [24]. Key aspects of the task force recommendations include the following:

●As in other patient populations, treatment choices in girls with epilepsy should be that of a shared decision between clinician and patients, based on a careful risk-benefit assessment of reasonable treatment options for the patient's seizure or epilepsy type.

●Given the risks associated with valproate exposure in utero, valproate should be avoided whenever possible as initial treatment of epilepsy in girls and women of childbearing potential.

●Valproate should generally be avoided for treatment of focal epilepsies, since multiple alternative, equally effective drugs are available that may have lower teratogenic risks.

●For cases in which valproate is considered the most effective option (eg, some genetic generalized epilepsies), valproate can still be considered as a first-line treatment option. In such cases, patients and caregivers should be informed of the risks associated with valproate use during pregnancy (see "Risks associated with epilepsy during pregnancy and the postpartum period", section on 'Valproate') as well as the relative risks and benefits of alternative treatment options. Options for effective contraception in the setting of antiseizure medication therapy should also be reviewed. (See 'Additional considerations in adolescent girls' below.)

●When valproate is used in girls and women of childbearing potential, it should be prescribed at the lowest effective dose and, when possible, at doses not exceeding 500 to 600 mg/day.

Additional considerations relevant to antiseizure medication therapy in women with epilepsy who are planning pregnancy or who become pregnant are reviewed separately. (See "Management of epilepsy during preconception, pregnancy, and the postpartum period", section on 'Preconception management'.)

Drug-drug interactions — Many commonly used drugs can alter the metabolism of antiseizure medications and vice versa (table 5A-C). Strong hepatic enzyme inducers (eg, phenytoin, phenobarbital, primidone, carbamazepine, oxcarbazepine, felbamate) will lower the levels of drugs metabolized in the liver, and liver enzyme inhibitors (eg, valproate) will slow the metabolism of the same drugs. Cimetidine, propoxyphene, erythromycin, fluoxetine, and clarithromycin are examples of enzyme inhibitors used in children that may elevate the serum levels of some antiseizure medications.

Many of the newer antiseizure medications are nonenzyme-inducing drugs and therefore have less potential for drug-drug interactions. Nonenzyme-inducing antiseizure medications such as levetiracetam are a common choice as first-line therapy when the avoidance of drug-drug interactions is critical, such as in children with tumor-associated epilepsy who are being treated with chemotherapy.

Specific interactions of antiseizure medications with other medications may be determined using the Lexicomp drug interactions tool.

INITIATION OF ANTISEIZURE MEDICATION THERAPY

Baseline laboratory evaluation — Routine laboratory tests are not necessary for many antiseizure medications; exceptions include cannabidiol, carbamazepine, felbamate, valproate, and others as listed in the tables (table 2A-B). (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects".)

Before initiating valproate in children under the age of three years, or in children who might have a neurometabolic disorder, we suggest obtaining serum ammonia, pyruvate, lactate, and carnitine levels and serum amino acid and urine organic acid analyses. Valproate should be avoided if there is clinical suspicion that the child's illness is progressive when no etiology has been determined. (See "Valproic acid poisoning", section on 'Hepatotoxicity'.)

Role of pretreatment HLA testing — Screening for the human leucocyte antigen (HLA)-B*1502 allele is recommended prior to starting carbamazepine or oxcarbazepine in patients with Asian ancestry. The HLA-A*31:01 and HLA-A*24:02 alleles are found in a broader range of ethnic groups and also increase the risk for hypersensitivity reactions [25-27]. However, there is no consensus about testing for these alleles.

The risk of carbamazepine-induced hypersensitivity reactions, including Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN), is increased in patients with HLA-B*1502 allele, which occurs almost exclusively in individuals with Asian ancestry. An increased risk of SJS or TEN in patients with the HLA-B*1502 allele has also been described for oxcarbazepine and phenytoin, although the magnitude of the risk appears to be lower than with carbamazepine. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Carbamazepine' and "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Oxcarbazepine' and "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Phenytoin and fosphenytoin' and "Stevens-Johnson syndrome and toxic epidermal necrolysis: Pathogenesis, clinical manifestations, and diagnosis", section on 'HLA polymorphism and pharmacogenetics'.)

Carbamazepine, oxcarbazepine, and phenytoin should be avoided in patients known to carry the HLA-B*1502 allele unless the estimated benefits clearly outweigh the risks.

Drug administration and dosing — The kinetics of an antiseizure medication determine the initial dose and the interval between dose increments during titration. Suggested dosing regimens, target serum levels, and laboratory testing for many commonly used antiseizure medications are listed in the table (table 2A-B).

Titration and maintenance dosing should be individualized:

●Most antiseizure medications should be started at approximately 10 to 25 percent of the planned maintenance dose. Antiseizure medications with a long half-life can be started at close to the maintenance dose. In general, the dose should be increased at intervals not exceeding five half-lives to allow the serum level to plateau between each dose increment. If seizures are frequent, the titration may be accelerated to achieve a therapeutic benefit, but this may come at the cost of increased side effects.

●The antiseizure medication dose should be increased until seizures stop, unremitting adverse effects occur, or serum levels reach a high or supratherapeutic range without a significant impact upon seizure frequency. The recommended upper therapeutic serum levels of most of the antiseizure medications can be exceeded if side effects are absent. This should be done with particular caution with phenytoin because of its nonlinear pharmacokinetics and with valproate because of dose-related thrombocytopenia (table 2A-B).

●If adverse effects appear with upward titration but are tolerable, the dose should remain stable for several weeks to determine if the symptoms remit. Dose increases can continue at a slower rate if side effects remit and seizures continue.

●Some patients may require greater-than-standard doses of an antiseizure medication to reach therapeutic levels. Low levels at high doses often are caused by poor compliance but also may be secondary to hypermetabolism of the antiseizure medication (ie, higher-than-normal clearance because of increased hepatic metabolism). If levels are not increasing as anticipated and noncompliance is unlikely, the child should be given a loading dose under medical observation and levels should be measured over the next 12 to 24 hours. A genetically regulated hypermetabolism probably exists if levels remain low, and higher daily doses are indicated.

●Occasionally, children manifest slow drug clearance, often caused by inherited variations in the degradation enzymes. These individuals require lower maintenance doses.

●Ideally, daily antiseizure medication doses should be given at intervals shorter than the half-life in order to avoid wide excursions in serum levels. The longer the half-life, the less frequent the dosing. Compliance is improved when medications are taken on schedule and prompted by an association with a routine daily activity, such as mealtime or brushing teeth. A variation of an hour or so in scheduled dosing is usually not significant. Children should not be awakened at night to take antiseizure medications.

●As a child grows, the antiseizure medication dose may need to be increased to keep up with his or her body mass. However, the clearance of some antiseizure medications decreases and approaches adult rates in adolescence. Serum levels may remain stable as the declining per kilogram dose is balanced by the declining clearance.

Additional considerations in adolescent girls — As noted above, the potential teratogenicity of various antiseizure medications should be discussed and weighed when choosing a specific antiseizure medication in girls, since in many cases initial therapy will be continued long term, into the childbearing years. (See 'Teratogenicity' above and "Risks associated with epilepsy during pregnancy and the postpartum period", section on 'Effects of ASMs on the fetus and child'.)

The importance of an effective method of contraception should also be reviewed with girls who are sexually active or may become sexually active while taking antiseizure medications. Certain antiseizure medications have the potential to lower the efficacy of hormonal contraception through hepatic enzyme induction (table 6). Conversely, hormonal contraception can lower serum levels of some antiseizure medications (eg, lamotrigine) [28]. The World Health Organization (WHO) suggests that individuals taking enzyme-inducing antiseizure medications or lamotrigine use a method of contraception other than hormonal pill, patch, or ring contraceptives [29]. Long-acting reversible contraceptive methods are an effective alternative in this setting. (See "Management of epilepsy during preconception, pregnancy, and the postpartum period", section on 'Contraception' and "Contraception: Counseling and selection".)

Folate supplementation is advised in adolescent girls on antiseizure medications and those with folate deficiency.

FOLLOW-UP AND MONITORING

Utility of serum drug levels — Serum antiseizure medication levels are informative for certain antiseizure medications but not others. Antiseizure medications for which serum drug levels can be useful are noted in the tables (table 2A-B). The availability of a serum drug level testing for an antiseizure medication does not always mean that levels are useful clinically. As an example, serum drug levels for levetiracetam can be ordered but have not been shown to correlate with clinical efficacy or tolerability [30].

Even for antiseizure medications where serum levels are useful, serum levels should not be used in isolation to guide to therapy. The therapeutic range is different for each patient. Many patients will achieve seizure control at levels below the recommended range; others require higher levels. There is no reason to increase the dose if seizures stop when the serum level is "low" or "subtherapeutic." If the level reaches the "therapeutic range," yet seizures continue, the level should be increased as long as there are no adverse effects.

It is helpful to obtain a baseline antiseizure medication level once seizure control has been attained, which can be compared with a level obtained when seizures recur to determine whether the breakthrough is caused by a low serum level. Serum levels may be low from poor compliance, interaction with another medication, decreased absorption (eg, during a diarrheal illness or gastritis with vomiting), or change in medication preparation (eg, brand name to generic drug formulation). If a child is placed on another long-term medication, measuring the antiseizure medication level after the dose of the new drug has been stabilized may be necessary.

Monitoring for specific drugs — Routine laboratory screening of hematologic and hepatic function is generally not necessary for asymptomatic children receiving antiseizure medications (table 2A-B) [31,32]. However, certain antiseizure medications require monitoring based on specific dose-related or idiosyncratic adverse effects, as listed below.

Children on long-term antiseizure medications, especially enzyme-inducing drugs such as phenytoin, carbamazepine, and phenobarbital, have an increased prevalence of hyperlipidemia, low folate, low vitamin B12, and hyperhomocysteinemia [33,34]. Screening for hyperlipidemia and folate deficiency should be considered in older children on enzyme-inducing antiseizure medications.

●Valproate – Although routine monitoring of hepatic function has not been shown to permit early identification of serious toxicity or improve outcome for patients taking valproate (valproic acid), many clinicians choose to obtain LFTs once or twice a year in patients who are clinically asymptomatic. The US Food and Drug Administration (FDA) recommends checking LFTs prior to initiating treatment and at frequent intervals thereafter, especially during the first six months. Hepatic enzyme elevations that are less than three times normal in an asymptomatic child are unlikely to be significant. Higher levels should be repeated in a few weeks and the medications stopped if levels are increasing rapidly or if the child becomes symptomatic.

Valproate is associated with a relatively high incidence of minor and usually insignificant elevations of one or more liver enzymes and serum ammonia. In a study of 206 adults and children taking anticonvulsants including phenytoin, carbamazepine, valproic acid, and phenobarbital, the serum gamma glutamyl transpeptidase (GGT) was elevated in 75 percent and alanine aminotransferase (ALT) in 25 percent [35]. Other researchers have confirmed these results [36]. Elevations in aspartate transaminase (AST) are recorded less commonly and may be a more specific marker of liver dysfunction.

Increases in plasma ammonia are common and may occur in the absence of abnormal liver function tests. Risk factors for hyperammonemia in children taking valproate include young age, increased valproate dose, low serum carnitine, and concomitant use of phenytoin, phenobarbital, carbamazepine, or a carbonic anhydrase inhibitor (eg, acetazolamide, topiramate, or zonisamide). (See "Valproic acid poisoning", section on 'Hyperammonemia'.)

Most cases of fatal hepatotoxicity with valproate have been in children younger than three years and usually occurred in the first six months of therapy [37]. These children often develop encephalopathies and are on polytherapy because of the severe nature of their seizures. It seems likely that many of these patients have an underlying metabolic disorder causing the seizures (eg, urea cycle defects, organic acidurias, mitochondrial cytopathies, storage diseases, or other progressive conditions such as Alpers syndrome), which are exacerbated by exposure to valproic acid. Also, in some cases, the liver enzymes remain normal despite advanced hepatic failure with nausea, vomiting, worsening of seizures, and encephalopathy.

●Carbamazepine – In practice, we typically obtain a CBC after the first month of carbamazepine therapy. If the white blood count (WBC) is significantly decreased, it is repeated every three to four weeks until the counts stabilize. If the absolute neutrophil count (ANC) falls below 800 to 1000, the medication should be stopped.

Leukopenia is not uncommon with carbamazepine, often appearing in the first two to three months of therapy [38]. Severe aplastic anemia or agranulocytosis is reported but rare, occurring in 2 per 575,000 exposures. These severe blood dyscrasias are more common in adults. A drop of the white count into the 3000 to 4000 range characterizes the more common benign leukopenia, which will usually gradually return toward normal or may remain at a mildly low level for the duration carbamazepine therapy.

●Oxcarbazepine – Hyponatremia has been reported with the use of oxcarbazepine, but this is very rare in children. The two predominant risk factors appear to be concomitant use of other "sodium-depleting" medications (eg, tricyclic antidepressants, thiazide diuretics, atypical antipsychotics) and excessive free water consumption. The effect is dose related and usually occurs within the first three months of therapy or following the addition of another sodium-depleting medication. It is not usually necessary to reduce or stop oxcarbazepine due to hyponatremia; the serum sodium level usually gradually returns to normal or remains at a mildly reduced level and is nearly always asymptomatic. For sodium levels less than 120 mEq/L or symptoms of hyponatremia, a dose reduction of oxcarbazepine or mild fluid restriction is usually successful. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Hyponatremia'.)

●Topiramate and zonisamide – Due to the risk of metabolic acidosis and nephrolithiasis, serum bicarbonate should be measured at baseline and monitored periodically thereafter in children treated with topiramate or zonisamide. Dose reduction or drug discontinuation should be considered in patients with persistent or severe metabolic acidosis. If the drug is continued, alkali therapy may be warranted as in type 2 (proximal) renal tubular acidosis. (See "Treatment of distal (type 1) and proximal (type 2) renal tubular acidosis".)

Topiramate and zonisamide are partial carbonic anhydrase inhibitors and may cause a mild to moderate chronic metabolic acidosis or nephrolithiasis in a minority of children [39-45]. As an example, a 2014 systematic review found that metabolic acidosis was present in 29 percent of patients on topiramate and in 7 percent of patients on zonisamide [45]. The risk and severity of acidosis are increased if there are other predispositions to metabolic acidosis, such as renal disease or concurrent use of a ketogenic diet [39,40]. (See "Ketogenic dietary therapies for the treatment of epilepsy" and "Seizures and epilepsy in children: Refractory seizures", section on 'The ketogenic diet'.)

Potential complications of chronic metabolic acidosis in children include impaired growth, rickets, or osteomalacia [46,47].

●Vigabatrin – Vigabatrin is associated with risks of permanent retinal dysfunction and concentric visual field constriction that have been noted as early as nine months after initiation of treatment [48]. The frequency of deficits increases with treatment duration and cumulative dose. Baseline and serial ophthalmologic evaluations are required in children treated with vigabatrin. In the United States, prescribers, patients, and pharmacies must participate in the Risk Evaluation and Mitigation Strategies (REMS) program. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Vigabatrin' and "Infantile epileptic spasms syndrome: Management and prognosis", section on 'Vigabatrin'.)

Monitoring for adverse events — The clinician should be vigilant and have a high index of suspicion for adverse events in any child receiving chronic antiseizure medications. A seemingly benign illness that lasts for more than a few days should prompt a complete blood count (CBC) and/or liver function studies. Vomiting (the most common early symptom of hepatotoxicity or pancreatitis), prolonged unexplained fever, easy bruising, extreme fatigue or lethargy, flu-like symptoms, unexplained worsening of seizures, change in mental status, and abdominal pain should lead to further investigations [49].

However, with the exception of felbamate, which is associated with a relatively high risk of aplastic anemia and requires close laboratory monitoring (table 2A-B), the risk of serious adverse effects with most of the antiseizure medications is low. In most instances, serious adverse events occur in the first few months after the antiseizure medication is initiated.

Note that many idiosyncratic reactions related to antiseizure medications, including Stevens-Johnson syndrome, toxic epidermal necrolysis, serum sickness reactions, and pancreatitis, are not predicted by presymptomatic blood test abnormalities; an exception is carbamazepine-induced hypersensitivity reactions related to certain human leucocyte antigen [HLA] types). (See 'Role of pretreatment HLA testing' above.)

A family history can be helpful in making decisions about laboratory testing. Adverse reactions to medications, particularly those that are hematologic or cutaneous, or a strong family history of autoimmune disorders, should heighten the concern for idiosyncratic reactions and hematologic complications. A family as well as personal history of renal stones should be sought when topiramate or zonisamide is considered [41,42]. (See "Nephrolithiasis in renal tubular acidosis".)

Bone health — Screening for vitamin D deficiency and supplementation with calcium and vitamin D are low-risk interventions that are suggested for children and adults on long-term antiseizure medication therapy. (See "Antiseizure medications and bone disease" and "Ketogenic dietary therapies for the treatment of epilepsy", section on 'Adverse effects' and "Epilepsy in children: Comorbidities, complications, and outcomes", section on 'Bone health'.)

Psychiatric and behavioral health screening — Increased suicidality has been linked to several antiseizure medications. Patients taking antiseizure medications should be monitored for emergence or worsening of suicidal ideation or depression. The International League Against Epilepsy (ILAE) endorses routine screening of cognition, mood, and behavior in new-onset epilepsy [50].

Routine screening can take the form of self-report questionnaires, computerized assessment batteries, and/or clinical questioning of mood, psychologic adjustment, and subjective cognitive complaints (eg, attention, memory, or word-finding difficulties). Formal neuropsychologic assessment may be considered when a focal cognitive deficit is suspected or apparent or when there is a question of neurodevelopmental delay, behavioral or learning difficulties, or cognitive decline. Serial neuropsychologic assessments can also be useful for evaluating the effects of the disorder and its treatment. (See "Epilepsy in children: Comorbidities, complications, and outcomes", section on 'Psychiatric and behavioral health'.)

ADHERENCE TO ANTISEIZURE MEDICATIONS

Nonadherence — Rates of nonadherence to prescribed antiseizure medication therapy are difficult to measure but are probably higher than is generally appreciated and can contribute to inadequate seizure control. One prospective observational study in 124 children (2 to 12 years old) with newly diagnosed epilepsy found that 58 percent demonstrated nonadherence during the first six months of treatment [51]. A pattern of nonadherence was often established within the first month. In a follow-up study of the same cohort, nonadherence during the first six months was associated with worse seizure control at four years [52].

The complexity of drug regimens and the occurrence of side effects are believed to contribute to nonadherence. Low socioeconomic status has also been identified as a risk factor [51].

Improving adherence — We advise measures to promote general medication adherence, such as daily pill boxes and automated pill dispensers to organize and keep track of medication dosing, and smartphone apps for patients and caregivers that send reminders when it is time to take medications and notifications if a dose has been missed.

Small studies have explored ways to improve upon adherence in patients with epilepsy. A systematic review of randomized trials testing the effectiveness of adherence interventions in adults (five trials) and children (one trial) with epilepsy found that behavioral interventions (eg, use of intensive reminders) were associated with somewhat better results than education and counseling [53].

REFRACTORY SEIZURES — Most children with epilepsy achieve reasonably good seizure control with antiseizure medication therapy, but some are refractory despite numerous medications. Medical treatment failure is often apparent early in the course of treatment. In these cases, referral to a comprehensive epilepsy center is appropriate to explore additional therapeutic options, including epilepsy surgery, vagus nerve stimulation, and the ketogenic diet. (See "Seizures and epilepsy in children: Refractory seizures".)

DURATION OF ANTISEIZURE MEDICATION THERAPY

When to consider withdrawal — Withdrawal of antiseizure medication therapy should be considered in most children who are seizure-free for at least 18 to 24 months, regardless of the etiology of the seizures [54]. Children with neurologic deficits may also be considered for antiseizure medication withdrawal if they have been seizure-free for an extended period of time, even though the presence of a motor or cognitive deficit is a risk factor for recurrence. (See 'Risk of seizure relapse after withdrawal' below.)

Method of withdrawal — Antiseizure medications should be tapered rather than halted abruptly. There are limited data to guide an antiseizure medication tapering schedule [55]. Rapid changes (over days to a few weeks) in drug treatment increase the risk of seizures. Slower rates of antiseizure medication taper, over several weeks to a few months, are generally recommended. In particular, benzodiazepines and barbiturates are associated with withdrawal seizures and should be discontinued very gradually.

Risk of seizure relapse after withdrawal — For patients who have been seizure free for a two-year period, the likelihood of seizure recurrence after stopping antiseizure medications is approximately 30 to 40 percent [56-58].

●Length of seizure-free period – A meta-analysis of five studies found that earlier discontinuation (before two years) of antiseizure medication therapy was associated with a higher risk for seizure relapse (risk ratio [RR] 1.34), particularly in children with focal epilepsy or an abnormal electroencephalogram (EEG) [59]. Longer seizure-free periods were associated with only a slightly lower incidence of recurrence, and therefore longer observation periods (ie, >2 to 3 years) are not warranted.

●Other risk factors – The following factors indicate an increased likelihood of recurrent seizures with discontinuation of antiseizure medications:

•Presence of a motor or cognitive deficit.

•Abnormal EEG at the time of discontinuation in children with epilepsy of unknown cause [60]. The predictive value of EEG in children with remote symptomatic epilepsy is not as clear.

•Certain acquired or genetic epilepsies [57,60]. Note that some genetic epilepsies nearly always remit (eg, childhood absence) but others are less likely to remit (eg, juvenile myoclonic epilepsy [JME]).

•Structural abnormality on brain magnetic resonance imaging (MRI).

•Short treatment period (6 to 12 months) prior to discontinuation [61,62].

Factors that may be predictive of good outcome include [60-63]:

•Younger age

•Having only absences as a seizure type

In one study of 65 children with cerebral palsy and epilepsy, antiseizure medication treatment was stopped after at least two seizure-free years; seizure relapse occurred in 42 percent [64]. Children with hemiplegia had a higher relapse rate (62 percent) than did those with a paraplegia (14 percent). This probably relates to the fact that the major pathology with paraplegia is in the subcortical white matter with minimal involvement of the more epileptogenic cortical neurons. In this study, the EEG at the time of discontinuation was not of value in predicting recurrence.

●Risk of intractable seizures – The risk of recurrent intractable seizures after discontinuing antiseizure medications in seizure-free children is very low. In a cohort study of 260 children who became seizure free and stopped antiseizure medications and were followed for four to five years, three children (1 percent) developed recurrent seizures that could not be controlled again with medication [65]. It is not clear that these recurrent seizures could have been prevented had antiseizure medications not been discontinued.

Recommendations of others — A 2021 practice advisory from the American Academy of Neurology (AAN) for antiseizure medication withdrawal in seizure-free patients reached the following conclusions for children [54]:

●There is probably no significant difference in the risk of seizure recurrence between those who begin ASM withdrawal after two years of seizure freedom compared with four years of seizure freedom. There is insufficient evidence of significant difference in the risk of seizure recurrence between those who begin ASM withdrawal after 18 months of seizure freedom compared with withdrawal after 24 months, based upon findings from a small randomized trial [66].

●Clinicians must account for the known natural history of specific electroclinical epilepsy syndromes when counseling about ASM withdrawal in children.

●Patients, families, and caregivers should be counseled about the following aspects of ASM withdrawal [54]:

•In children who are seizure-free for at least 18 to 24 months and who do not have an electroclinical epilepsy syndrome that precludes ASM withdrawal, clinicians should discuss the risks and benefits of ASM withdrawal. The discussion should specifically include and document that if seizures recur during or after withdrawal, there is a small chance they will no longer respond to medication.

•Withdrawal of ASMs does not clearly increase risk of seizure recurrence.

•Recurrent seizures put children at risk for status epilepticus and death. However, existing data do not suggest an increased risk of status epilepticus or death after ASM withdrawal.

●An EEG should be obtained if there is agreement among the physician, patient, and family (or guardians if involved) to pursue consideration of ASM withdrawal for a child who is seizure-free for at least 18 to 24 months [54]. If the EEG does not show epileptiform activity, ASM withdrawal should be offered, at a rate no faster than 25 percent every 10 to 14 days.

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".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Epilepsy in children (The Basics)")

●Beyond the Basics topic (see "Patient education: Treatment of seizures in children (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●When to start antiseizure medication therapy

•First unprovoked seizure – In most cases, we refrain from starting antiseizure medication therapy for a child with a first unprovoked seizure, which is a seizure of unknown etiology as well as one that occurs in relation to a preexisting brain lesion or progressive nervous system disorder. However, antiseizure medication treatment may be warranted after a first unprovoked seizure for select children with specific epilepsy syndromes or those with a high risk of seizure recurrence, including those with a potential remote symptomatic etiology, particularly if the seizure was focal and the electroencephalogram (EEG) or brain magnetic resonance imaging (MRI) is abnormal. (See 'First unprovoked seizure' above.)

•Second unprovoked seizure – We start antiseizure medication therapy for most children who present with a second unprovoked seizure, since seizure recurrence indicates that the patient has a substantially increased risk for additional seizures (ie, epilepsy). (See 'Second unprovoked seizure' above.)

•Acute symptomatic (provoked) seizure – Long-term antiseizure medication therapy is generally not warranted for children who have a seizure in the setting of an acute illness or brain insult (eg, acute infection, acute head injury) since they have a low risk of seizure recurrence (See 'Acute symptomatic seizure' above.)

•Febrile seizure – Given the benign nature of febrile seizures, the risks of side effects from antiseizure medications generally outweigh the benefits for most patients. This is discussed in more detail separately. (See "Treatment and prognosis of febrile seizures", section on 'Role of preventive therapy'.)

●Selection of antiseizure medication – The antiseizure medication chosen for initial therapy should be one that is effective for a particular seizure type or syndrome and that is safe and well tolerated. Other considerations include dose formulation, dose frequency, the relative risk of certain adverse effects, and the potential for drug-drug interactions. Suggested dosing regimens, target serum levels, and laboratory testing for many commonly used antiseizure medications are listed in the table (table 2A-B). (See 'Selection of an antiseizure medication' above and 'Initiation of antiseizure medication therapy' above.)

●Follow-up and monitoring

•For specific medications – Serum antiseizure medication levels are informative for certain antiseizure medications but not others, and some antiseizure medications (eg, valproate, carbamazepine, oxcarbazepine, topiramate, zonisamide, and vigabatrin) require monitoring based on specific dose-related or idiosyncratic adverse effects, as noted in the table (table 2A-B). (See 'Follow-up and monitoring' above.)

•For worrisome adverse events – Vigilance for adverse events is warranted for any child receiving chronic antiseizure medications (table 3 and table 4). A seemingly benign illness that lasts for more than a few days should prompt a complete blood count (CBC) and/or liver function studies. Vomiting (the most common early symptom of hepatotoxicity or pancreatitis), prolonged unexplained fever, easy bruising, extreme fatigue or lethargy, flu-like symptoms, unexplained worsening of seizures, change in mental status, and abdominal pain should lead to further investigations. (See 'Monitoring for adverse events' above.)

•For depression and suicidality – Increased suicidality has been linked to several antiseizure medications. Patients taking antiseizure medications should be monitored for emergence or worsening of suicidal ideation or depression. (See 'Psychiatric and behavioral health screening' above.)

●Treatment failure – Most children with epilepsy achieve reasonably good seizure control with antiseizure medication monotherapy or polytherapy, but some are refractory despite numerous medications. Medical treatment failure is often apparent early in the course of treatment. In these cases, referral to a comprehensive epilepsy center is appropriate to explore additional therapeutic options, including epilepsy surgery, vagus nerve stimulation, and the ketogenic diet. (See "Seizures and epilepsy in children: Refractory seizures".)

●When to consider medication withdrawal – Withdrawal of antiseizure medication therapy should be considered in most children after 18 to 24 months without seizures, regardless of the etiology of the seizures. The likelihood of recurrence after a two-year period without seizures is approximately 30 to 40 percent. Antiseizure medications should be tapered rather than halted abruptly. (See 'Duration of antiseizure medication therapy' above.)