Approach

The goal of treatment for any epilepsy syndrome is complete freedom from seizures. At the same time, the risk of adverse medicine effects need to be considered. Most treatment choices are based on expert opinion, as there is minimal good evidence.[40][41] A large randomised prospective trial comparing ethosuximide, lamotrigine, and valproate for the treatment of childhood absence epilepsy (CAE) concluded that ethosuximide might represent first-line treatment for CAE.[42] Initial treatment effect persisted at 12 month follow-up.[43]

Monotherapy is preferred, but an adjunctive medicine to first-line therapy may be required. Evidence suggests that earlier age of onset and male sex may increase the need for a second agent for seizure control.[44]

Typical absence seizures without a history of generalised tonic-clonic seizures (childhood absence epilepsy)

A syndrome with only typical absence seizures is likely to respond to ethosuximide, valproic acid, or lamotrigine as first-line treatments. Evidence suggests that ethosuximide and valproate have significantly greater efficacy than lamotrigine.[42] Ethosuximide had a small but significantly lower rate of attentional difficulties than valproate, suggesting that ethosuximide should be considered first-line treatment for CAE.[42] A Cochrane review concluded that ethosuximide is the optimal initial empirical monotherapy for children and adolescents with absence seizures.[41] Second-line agents include topiramate, zonisamide, and levetiracetam.

In the subgroup of patients with GLUT1 deficiency, a ketogenic diet is recommended. Patients are monitored and treated by an epileptologist. This is typically a high-fat, adequate-protein, low-carbohydrate diet, and should be initiated in hospital, under close medical supervision. It may take a couple of months before a clinical response is noted. Antiepileptic medication is continued initially. If a patient responds very well and has little or no seizure activity, medication is tapered slowly.

Typical absence seizures with a history of generalised tonic-clonic seizures (CAE, JAE, JME)

If there is any history of generalised tonic-clonic seizures (childhood absence epilepsy [CAE], juvenile absence epilepsy [JAE], and juvenile myoclonic epilepsy [JME]), ethosuximide is less appropriate, and valproic acid and lamotrigine would be preferred first-line agents. Second-line agents would include topiramate, zonisamide, and levetiracetam. Typically, second-line agents are added as adjunct therapy to first-line therapy. However, second-line therapy can be substituted for the first-line therapy, with the first-line medicine being weaned.

Atypical absence seizures

Valproic acid, lamotrigine, and topiramate are all indicated for first-line treatment of atypical absence seizures, syndromes with generalised epilepsies, or multiple seizure types. Typically, zonisamide and levetiracetam are second-line agents that are added as adjunct therapy to first-line therapy. However, second-line therapy can be substituted for the first-line therapy, with the first-line medicine being weaned.

Failure of therapy

Multiple other therapies can be considered if first and second line therapies have failed (i.e., lack of seizure freedom), such as acetazolamide, felbamate, the ketogenic diet, and vagal nerve stimulation. These are beyond the scope of this review and would be initiated by an epileptologist.

GLUT1 testing should be considered before initiating the ketogenic diet.

Drugs usually more appropriate for focal seizures, such as carbamazepine and phenytoin, are generally felt to worsen generalised seizures, including absence seizures.

Medication used

Ethosuximide

  • One double-blind RCT compared ethosuximide, valproic acid, and lamotrigine as first-line treatments in children with newly diagnosed childhood absence epilepsy. Ethosuximide and valproic acid had similar efficacy (53% and 58%, respectively, P = 0.35), but ethosuximide was better tolerated with fewer adverse attentional effects.[42] Initial treatment effect persisted at 12 month follow-up.[43] A Cochrane review concluded that ethosuximide is the optimal initial empirical monotherapy for children and adolescents with absence seizures.[41] Ethosuximide is considered very effective in patients with only typical absence seizures. It is generally well tolerated. A common adverse effect is gastrointestinal upset. In rare instances, it can cause aplastic anaemia, and hepatic or renal failure.

Valproic acid

  • Atypical absence seizures respond well to valproic acid.[45][46]

  • Valproic acid has been reported to be equally efficacious to ethosuximide in the treatment of absence seizures.[42][47][48][49] However, in a double-blind RCT of children with newly diagnosed childhood absence epilepsy, valproic acid was associated with increased risk of adverse attentional effects compared with ethosuximide.[42]

  • Valproic acid is reported to be effective treatment in juvenile absence epilepsy and juvenile myoclonic epilepsy.[50][51]

  • Medicines containing valproate increase the risk of congenital malformations and developmental problems in the infant/child if taken during pregnancy (see 'Safety of anticonvulsants in pregnancy').

Lamotrigine

  • In open label and crossover studies, lamotrigine appeared to be as effective as valproic acid for typical absence seizures in children, and generalised epilepsy.[52][53][54][55][56] However, in a double-blind, randomised controlled trial of children with newly diagnosed childhood absence epilepsy, ethosuximide and valproate were significantly more likely to be effective than lamotrigine.[42]

  • Lamotrigine has been shown to be of benefit in juvenile myoclonic epilepsy and for some seizure types of Lennox-Gastaut syndrome.[54][57][58]

Topiramate

  • There are good data for the use of topiramate for primary generalised tonic-clonic seizures but not for absence seizures.[54][59]

  • Topiramate has been shown to have some efficacy in Lennox-Gastaut syndrome as adjunctive therapy.[60] It can also be used as monotherapy.

Zonisamide

  • Small case series and abstracts have suggested efficacy of zonisamide in reducing seizure frequency in patients with typical absence seizures, as well as refractory primary generalised epilepsy.[54] One retrospective chart review of 45 patients aged 18 years or under with absence seizures found a 51.1% rate of seizure elimination with zonisamide.[61]

Levetiracetam

  • Levetiracetam is indicated as adjunctive therapy for juvenile myoclonic epilepsy (JME).[40][54][62] A review concluded that levetiracetam is an effective adjunct in patients with insufficiently controlled juvenile absence epilepsy (JAE) and JME.[63]

  • One small, prospective study (n=21) suggested that levetiracetam monotherapy may be effective in patients with CAE and JAE.[64] However, a randomised placebo-controlled trial conducted in children with newly diagnosed CAE or JAE for 2 weeks reported a 23.7% response rate to levetiracetam monotherapy, which was not significantly higher than in the placebo arm.[65] Of note, the trial was of short duration to minimise exposure to placebo, and high levetiracetam doses could not be attained.

Safety of anticonvulsants in pregnancy

For women and girls of childbearing potential, the safety of anticonvulsants in pregnancy must be taken into account in choice of medication.

In both the US and Europe, valproate and its analogues are contraindicated during pregnancy due to the risk of congenital malformations and developmental problems in the child. If it is not possible to stop valproate, treatment may be continued with appropriate specialist care. Valproate and its analogues must not be used in female patients of childbearing potential unless there is a pregnancy prevention programme in place and certain conditions are met.[66] If the patient is taking the drug to prevent major seizures and is planning to become pregnant, the decision of continuing valproate versus changing to an alternate agent should be made on an individual basis.

A review of the safety of anticonvulsants (other than valproate) in pregnancy by the UK Medicines and Healthcare products Regulatory Agency concluded that lamotrigine and levetiracetam, at maintenance doses, are not associated with an increased risk of major congenital malformations. Available studies also do not suggest an increased risk of neurodevelopmental disorders or delay associated with in-utero exposure to lamotrigine or levetiracetam, but data are more limited. Data for other drugs show an increased risk of major congenital malformations associated with topiramate; and an increased risk of fetal growth restriction associated with topiramate and zonisamide.[67] A specialist should be consulted for more guidance on the use of specific drugs in pregnancy.

Drug discontinuation

Seizure freedom for long periods of time can occur with anticonvulsant therapy or after surgical treatment. Patients taking anticonvulsants who achieve seizure freedom may eventually wish to discontinue their medication to avoid the adverse effects, psychological implications, and cost of ongoing treatment.

There is no statistically significant evidence to guide the timing of anticonvulsant discontinuation in adults. For adults who have been seizure-free for at least 2 years, clinicians should discuss the risks and benefits of medication discontinuation with the patient, including the risks of seizure recurrence and treatment resistance. Individual patient characteristics and preferences should be taken into account. Patients who are seizure-free after epilepsy surgery and are considering medication discontinuation should be informed that the risk of seizure occurrence is uncertain due to lack of evidence.[68] Abrupt medication discontinuation is inadvisable, but, beyond this, there is little evidence to guide the speed of medication taper in adults.[69]

For children who have been seizure-free for at least 18-24 months, and who do not have an electroclinical syndrome suggesting otherwise, discontinuation of anticonvulsant medication may be considered, as this does not clearly increase risk of seizure recurrence. The risks and benefits of discontinuation should be discussed with the patient and family. Provided that an EEG does not show epileptiform activity, discontinuation should be offered at a rate no faster than 25% every 10-14 days.[68]

Use of this content is subject to our disclaimer