Clinical aspects

Description

Treatment approach. The majority of newly diagnosed patients will have a good prognosis and become seizure-free with the first or second monotherapy, often requiring no more than modest or moderate medication doses (52; 06; 57; 14). They should be started on a single medication at low dose with increments over a number of weeks to establish an effective and tolerable regimen. Further dose adjustment can be made according to clinical response. An alternative medication is unavoidable when the patient develops intolerable adverse events. When and how to combine antiseizure medications remains controversial. Many authorities agree that it would be appropriate to consider combination therapy after two monotherapy attempts because the chance of success with a third medication is slim (17; 53; 07; 57; 31). Failure of two treatment schedules should prompt referral to a specialist center for further evaluation, including video-EEG telemetry to confirm the seizure or syndrome classifications, optimization of pharmacotherapy, and consideration of other therapeutic options, particularly epilepsy surgery.

Before starting prophylactic antiseizure medication in a patient with newly diagnosed seizures, a physician should be confident of the following:

All these should be thoroughly explained to the patient or guardian, and it should be ensured they fully understand.

Medication selection. Not only do antiseizure medications differ in the way they work, they vary substantially in their efficacy and side-effect profiles. The ideal profile of a first­choice antiseizure medication in the prophylactic treatment of epilepsies is determined by a number of factors (Table 2). The importance of these factors varies significantly according to whether the antiseizure medication is used in monotherapy or polytherapy. A first-choice medication should have a spectrum of activity and side-effect and interaction profiles that has the potential to produce seizure freedom without intolerable toxicity and long-term sequelae for that individual. The choice should be matched to the patient's seizure type or epilepsy syndrome, age, gender, weight, psychiatric history, other disease states, concomitant medication, and lifestyle (16; 57). Therefore, this decision should be based on a thorough clinical history and physical examination aided by laboratory findings, including the EEG and brain imaging. The “prescribing information” (Food and Drug Administration in the United States) and the “summary of product characteristics” (European Medicines Agency in the European Union) are the most complete single sources of information on a medication. The summary of package characteristics can be obtained in any European language from http://www.ema.europa.eu/ema. In United Kingdom these are also available from http://emc. medicines.org.uk.

Table 2. Factors to Consider When Prioritizing Antiseizure Medication Use

The antiseizure medication indications for the treatment of epileptic seizures according to the Food and Drug Administration package inserts are as follows. Please note that these indications are not the same with those of the European Medicines Agency and are often significantly different from the prevailing clinical practice (Table 3).

Brivaracetam. Treatment of partial onset seizures in patients 16 years of age and older with epilepsy.

Cannabidiol. Adjunctive treatment of seizures associated with two rare and severe forms of epilepsy: Lennox-Gastaut and Dravet syndrome, in patients aged 2 and older.

Carbamazepine

Cenobamate. Treatment of partial-onset seizures in adults, either as monotherapy or as adjunctive therapy.

Clobazam. Adjunctive treatment of seizures associated with Lennox-Gastaut syndrome in patients 2 years of age or older.

Clonazepam. Adjunct in the treatment of the Lennox-Gastaut syndrome (petit mal variant) as well as akinetic and myoclonic seizures. In patients with absence seizures (petit mal) who have failed to respond to succinimides, clonazepam may be useful.

Divalproex sodium. Monotherapy and adjunctive therapy in the treatment of adult patients and pediatric patients as young as 10 years of age with:

Eslicarbazepine acetate. Treatment of partial-onset seizures in patients 4 years of age and older.

Ethosuximide. Treatment of absence (petit mal) epilepsy.

Everolimus. Adjunctive treatment of adult and pediatric patients aged 2 and older with tuberous sclerosis complex–associated partial-onset seizures.

Felbamate. Felbamate is not indicated as a first-line antiseizure treatment. Felbamate is recommended for use only in those patients who respond inadequately to alternative treatments and whose epilepsy is so severe that a substantial risk of aplastic anemia or liver failure is deemed acceptable in light of the benefits conferred by its use.

If these criteria are met and the patient has been fully advised of the risk and has provided written acknowledgement, felbamate can be considered for either monotherapy or adjunctive therapy in the treatment of partial seizures, with and without generalization, in adults with epilepsy and as adjunctive therapy in the treatment of partial and generalized seizures associated with Lennox-Gastaut syndrome in children.

Fenfluramine. Treatment of seizures associated with Dravet syndrome in patients aged 2 and older.

Gabapentin. Adjunctive therapy in the treatment of:

Ganaxolone. Treatment of seizures associated with cyclin-dependent kinase-like 5 deficiency disorder in patients 2 years of age and older.

Lacosamide. Monotherapy or adjunctive therapy in patients 17 years and older with partial-onset seizures as adjunctive therapy. Injection for intravenous use is an alternative when oral administration is temporarily not feasible.

Lamotrigine. Adjunctive therapy for the following seizure types in patients 2 years of age or older: partial seizures, primary generalized tonic-clonic seizures, generalized seizures of Lennox-Gastaut syndrome.

Conversion to monotherapy in adults (16 years of age and older) with partial seizures who are receiving treatment with carbamazepine, phenytoin, phenobarbital, primidone, or valproate as the single antiseizure medication.

Safety and effectiveness of lamotrigine tablets have not been established (1) as initial monotherapy; (2) for conversion to monotherapy from antiseizure medications other than carbamazepine, phenytoin, phenobarbital, primidone, or valproate; or (3) for simultaneous conversion to monotherapy from two or more concomitant antiseizure medications.

Levetiracetam. Adjunctive therapy in the treatment of:

Oxcarbazepine.

Perampanel.

Phenobarbital. Treatment of generalized and partial seizures.

Phenytoin. Treatment of tonic-clonic (grand mal) and psychomotor (temporal lobe) seizures and prevention and treatment of seizures occurring during or following neurosurgery.

Pregabalin. Adjunctive therapy for adult patients with partial onset seizures.

Primidone. Monotherapy or adjunctive treatment of grand mal, psychomotor, and focal epileptic seizures. It may control grand mal seizures refractory to other anticonvulsant therapy.

Rufinamide. Adjunctive treatment of seizures associated with Lennox-Gastaut syndrome in pediatric patients 1 year of age and older and in adults.

Stiripentol. Adjunctive treatment of seizures associated with Dravet syndrome in patients aged 2 and older who are taking clobazam.

Tiagabine hydrochloride. Adjunctive therapy in adults and children 12 years and older in the treatment of partial seizures.

Topiramate.

Valproic acid.

Vigabatrin.

Zonisamide. Adjunctive therapy in the treatment of partial seizures in adults with epilepsy.

Table 3. Efficacy of Antiseizure Medications in Seizure Types

Table 4. Dosing Guidelines for Antiseizure Medication Monotherapy in Adolescents and Adults

Efficacy and tolerability. The profile of activity against different seizure types varies among the antiseizure medications (Table 3). In addition, certain epilepsy syndromes have been found to be particularly responsive to specific antiseizure medications. For instance, juvenile myoclonic epilepsy responds well to valproate and levetiracetam, whereas vigabatrin is regarded by many as the medication of choice for infantile spasms. A double-blind randomized controlled trial showed that ethosuximide and valproic acid were more effective than lamotrigine for childhood absence epilepsy (29). Therefore, it is of paramount importance to accurately classify the seizure type and epilepsy syndrome and choose the most appropriate medication for that individual.

Systematic reviews of randomized, controlled trials comparing phenobarbital, primidone, phenytoin, carbamazepine, and valproate for partial and tonic-clonic seizures identified no significant difference in efficacy among these established antiseizure medications, with the possible exception of a small benefit of carbamazepine over valproate for partial seizures (46). Phenobarbital and primidone produced higher withdrawal rates due to their sedative effect, although these medications are often well tolerated at moderate dosing in clinical practice.

None of the newer antiseizure medications have demonstrated superior efficacy when tested against the established agents in monotherapy studies for the treatment of newly diagnosed partial and tonic-clonic seizures. Evidence of better tolerability has been shown in some cases, in particular fewer neurotoxic side effects. Thus, lamotrigine and oxcarbazepine showed better overall effectiveness than carbamazepine and phenytoin, respectively.

Based on the findings from randomized, controlled trials, various national and international guidelines have been developed for the use of antiseizure medications as initial monotherapy for epilepsy. These guidelines focus on the efficacy and effectiveness aspects in the process of medication selection (25; 28; 55).

These guidelines are valuable in providing a working framework for the treatment of new-onset epilepsy. The differences in their recommendations may reflect fundamental differences in the purposes of the guidelines and in the approaches adopted by the groups developing them. It is important to bear in mind that these treatment guidelines focus on short-term efficacy and effectiveness evidence alone, and other patient-related factors, discussed below, must be considered when choosing the most appropriate treatment for the individual patient. Continuous advances in medication development also mean that these guidelines become rapidly outdated.

Table 5. Pharmacological Properties of Antiseizure Medications

Polytherapy. Given that most patients respond to their first or second antiseizure medication, combination therapy should usually be considered after failure of two monotherapy regimens due to lack of efficacy. If poor tolerability is the problem, a third monotherapy could reasonably be tried. Some patients experiencing many seizures immediately before diagnosis may be best treated with combination therapy following a partial response to the first antiseizure medication, particularly if an underlying anatomical problem, such as cortical dysplasia or hippocampal atrophy, has been identified (38).

When making a decision about polytherapy, one should first scrutinize the possible or probable reasons why the monotherapy failed. The following possibilities, which often require re­evaluation of diagnosis (genuine epileptic seizures? what type of seizures?), would be considered (57):

The ideal profile of an antiseizure medication for polypharmacy includes all the factors that are important for monotherapy (Table 2), but with particular emphasis on the following points (57):

Strength of efficacy. This may be increased or weakened by pharmacokinetic and pharmacodynamic interactions.

Safety and tolerability. These are often worsened by pharmacokinetic and pharmacodynamic interactions.

Medication interactions. Interactions with other antiseizure medications, whether pharmacokinetic, pharmacodynamic, or both, are particularly unwanted in polytherapy. Raising the levels of concomitant antiseizure medications and pharmacodynamic interactions may lead to toxic effects. Conversely, decreasing their levels may increase and worsen seizures causing a vicious cycle in clinical management. With the exception of lacosamide, levetiracetam, and gabapentin, most of the newer antiseizure medications exhibit sometimes complex, undesirable, medication-to-medication interactions. For example, consider that lamotrigine:

Different mechanisms of action in relation to other concurrent antiseizure medications (Table 8). Antiseizure medication-to-medication interactions may be purely additive, antagonistic, or synergistic. Antiseizure medications with the same mechanism of action would be expected to be additive, whereas combining antiseizure medications with different mechanisms of action may have synergistic effects. A sodium channel blocker antiseizure medication combined with another that increases the GABA­ergic neurotransmission or that has multiple mechanisms is generally more effective than a combination of two sodium channel blockers (21; 17; 39). An antiseizure medication is unlikely to have better success and more likely to have additive adverse medication reactions if added to another antiseizure medication with the same mechanism of action.

The risks of polytherapy include:

Converting from polytherapy to monotherapy. Evidence from studies with older and newer antiseizure medications shows that a significant number of patients can be converted successfully from polytherapy to monotherapy without losing seizure control and, in some cases, with improved seizure control. In these cases, the antiseizure medication that appears, after careful consideration, to be the least effective is gradually withdrawn. “Gradually” sometimes means in steps of weeks or months. This should be particularly slow for certain antiseizure medications, such as phenobarbital and benzodiazepines, in order to avoid withdrawal seizures.

Medication-to-medication interactions. Antiseizure medication interactions are an important consideration in the treatment of epilepsy and, indeed, can be a major therapeutic challenge. The principal pharmacokinetic interaction relates to hepatic enzyme induction or inhibition whilst pharmacodynamic interactions principally entail adverse effect synergism, although examples of anticonvulsant synergism also exist. The older antiseizure medications are commonly associated with pharmacokinetic interactions by inducing the synthesis of a range of monooxygenase and conjugating enzymes (phenobarbital, primidone, phenytoin, and carbamazepine, in particular, accelerate the breakdown of many commonly prescribed lipid-soluble medications, including oral contraceptives and warfarin). Valproate is a weak enzyme inhibitor, which can slow the clearance of other antiseizure medications, such as phenytoin and lamotrigine. Hepatic enzyme induction stimulates the production and increases the amount of CYP enzymes. This, in turn, increases the rate of metabolism of medications metabolized by the CYP enzymes, thus, resulting in lower plasma concentrations. Conversely, hepatic enzyme inducers may increase the bioavailability of medications that require metabolism for their activation. The effect of hepatic induction, unlike hepatic inhibition persists for several days following the withdrawal of the enzyme-inducing medication.

Some of the new antiseizure medications are less interacting primarily because many are renally excreted or not hepatically metabolized (eg, gabapentin, lacosamide, levetiracetam, topiramate, vigabatrin) and most do not (or minimally) induce or inhibit hepatic metabolism. The least pharmacokinetic interactions are associated with gabapentin, lacosamide, tiagabine, vigabatrin, and zonisamide, whereas lamotrigine, felbamate, oxcarbazepine, and rufinamide are associated with the most. To date, felbamate, gabapentin, oxcarbazepine, perampanel, pregabalin, rufinamide, stiripentol, and zonisamide have not been associated with any pharmacodynamic interactions (58).

Pharmacodynamics refers to the biochemical and physiological effects of medications and their mechanisms of action. They may be additive, synergistic, or antagonistic. They may be beneficial, detrimental or both, as exemplified by lamotrigine combined with valproate (increased therapeutic efficacy but also increased risk of adverse medication reactions and teratogenicity).

Therapeutic medication monitoring. Monitoring of circulating concentrations of antiseizure medications became possible in the late 1960s, initially as a research tool but was rapidly adopted for widespread clinical use largely because phenytoin was shown to undergo saturation kinetics (63). It is important to bear in mind that the commonly quoted “therapeutic” or “reference” range has a probabilistic meaning; that is, it refers to the serum concentrations at which the majority of patients can be expected to show a response without undue adverse effects. Some patients will do well with medication concentrations below the lower limit of the range whereas others will tolerate amounts exceeding the upper level with benefit and without toxicity (59; 42). Therefore, the dose should not be increased or reduced just to attain a defined “therapeutic” concentration in patients who are seizure- and side-effect-free.

The ILAE position paper of 2008 is a practice guideline for therapeutic medication monitoring and is highly recommended for further reading (59). It also provides pharmacokinetic details, interactions with other antiseizure medications, and reference ranges for each antiseizure medication and discusses the role of therapeutic medication monitoring in children and the elderly and during pregnancy.

Serum medication monitoring is best performed with a specific clinical question in mind. Situations in which serum medication monitoring are most likely to be of benefit include: (1) when a person has attained the desired clinical outcome, to establish an individual therapeutic concentration that can be used at subsequent times to assess potential causes for a change in medication response; (2) as an aid in the diagnosis of clinical toxicity; (3) to assess compliance; (4) to guide dosage adjustment in situations associated with increased pharmacokinetic variability (eg, children, the elderly, patients with associated diseases, medication formulation changes); (5) when a potentially important pharmacokinetic change is anticipated (eg, in pregnancy or when an interacting medication is added or removed); (6) to guide dose adjustments for antiseizure medications with dose-dependent pharmacokinetics, particularly phenytoin (59).

Monitoring the circulating concentration is of greatest value with agents that exhibit nonlinear pharmacokinetics, in particular phenytoin. Increases and decreases in dosing may result in disproportionately large changes in serum concentration. Substantial variation in carbamazepine levels over the course of a day--as much as 100% with twice-daily dosing--makes interpretation problematic unless the timing of dosing and blood sampling are standardized. Because valproate does not exhibit a clear-cut concentration-effect-toxicity relationship, monitoring of this medication is not very helpful. Concentration-response-toxicity studies with the newer antiseizure medications have not demonstrated consistent correlations between circulating levels and efficacy or side effects (32).

Trough antiseizure medication plasma levels are important with regard to efficacy, whereas peak antiseizure medication plasma levels are important with regard to toxicity.

Prescription of brand-to-generic or generic-to-generic antiseizure medication. There is considerable concern among physicians and patients about the efficacy and safety of brand-to-generic antiseizure medication substitution or vice versa and also from one to another generic medication of different manufacturers (50). These concerns have been reasonably raised because of well-documented breakthrough seizures, emergency-treated epilepsy events, and toxicity as a result of such substitutions.

Substitution from one to another antiseizure medication of the same active substance is safe only when these are of therapeutic equivalence; medication products are considered to be therapeutic equivalents if they can be expected to have the same clinical effect and safety profile when administered to patients under the conditions specified in the labelling. However, even when the generic product is from a reliable source (something that is beyond the role of prescribing physicians) there may be problems with their therapeutic equivalence, particularly in terms of bioavailability and pharmacokinetic profiles.

Antiseizure medications are particularly vulnerable in these changes because many have a narrow reference concentration index between medication levels that are therapeutic and those that may cause adverse medication reactions.

“For antiepileptic drugs, small variations in concentrations between name brands and their generic equivalents can cause toxic effects or seizures when taken by patients with epilepsy. The American Academy of Neurology believes that the authorities should give complete physician autonomy in prescribing antiepileptic drugs” (50). It is because of these problems that formal recommendations worldwide uniformly warn against such substitutions of antiseizure medications.

A patient who is stabilized with an antiseizure medication in terms of seizure control and minimal adverse medication reactions should remain in the same brand or the particular generic product; interchanging products introduces risk.

When an antiseizure medication is first prescribed, it can be a brand name or a generic product providing that the latter has been truly tested in regard to the credibility of its manufacturers and bioequivalence with the brand product. Titration and maintenance should be with the same antiseizure medication product.

It is unreasonable to switch from one to another antiseizure medication, either from an expensive branded antiseizure medication to a cheaper bioequivalent generic product or vice versa. Switching is unnecessary and may impose significant risk to the patient. When such substitution is attempted, the patient should be well informed of possible consequences, such as seizure relapse or adverse medication reactions.

Prescriptions should clearly indicate the type of antiseizure medication formulation to be used, even for generic products (20; 57; 61). However, generic antiseizure medications with Food and Drug Administration–validated bioequivalence appears to be a safe clinical choice (33; 62).

Indications

Depending on study methodology and inclusion criteria, the chance of recurrence after a single unprovoked seizure reported in the literature ranges from 23% to 71% (09). Because a substantial proportion of patients will not have further episodes, and randomized studies suggest that early treatment does not improve prognosis (54; 45), most specialists do not recommend antiseizure medication therapy after a single seizure. However, when the chance of recurrence is high, eg, in the presence of an underlying cerebral lesion or epileptiform discharge on EEG, treatment may commence after the first tonic-clonic seizure.

Patients reporting more than one well-documented or witnessed seizure usually require treatment. Exceptions can include widely separated events, provoked seizures (eg, concomitant illness, photosensitive epilepsy, alcohol withdrawal), and benign childhood epilepsy syndromes. In addition, there is little chance of successful treatment in patients such as alcohol abusers, drug addicts, and conscientious objectors who are unlikely or unwilling to take medication. The decision of whether or not to start treatment should be made after ample discussion with the patient and his or her family of the risks and benefits of both courses of action.

Contraindications

Patients whose seizures are nonepileptic, including psychogenic/functional or “pseudoseizures,” should not be treated with antiseizure medications. Those whose seizures can be prevented by avoidance of provoking factors such as alcohol, illicit drugs, sleep deprivation, or flashing lights may not require long-term antiseizure medication therapy.

An antiseizure medication is contraindicated not only when it exaggerates seizures but also when it is ineffective in controlling the seizures it is supposed to treat. It may cause unnecessary adverse medication reactions and deprive the patient of the therapeutic effect that could be provided by another appropriate antiseizure medication (57).

Outcomes

Treatment with an antiseizure medication may fail due to lack of efficacy, intolerable adverse effects, or both. In the Glasgow series of 780 newly diagnosed patients, 47% became seizure-free with the first monotherapy, over 80% of whom required no more than modest to moderate doses. Another 10% responded to the second monotherapy. Only 2.3% of the cohort entered remission with the third monotherapy or with polytherapy (52; 53). Results from this and other studies suggest that failure of the first medication due to lack of efficacy is a powerful predictor of subsequent refractoriness and that the probability of remission after the failure of two to three successive medication regimens is slim (35; 53).

When the newer antiseizure medications were studied as add-on treatment in randomized, controlled trials in refractory epilepsy, a reduction in seizure frequency by 50% from the baseline was achieved in only 30% to 50% of patients. The clinical relevance of a 50% reduction in seizure frequency has been called into question (51). Studies including patients treated surgically (27) or medically (12) suggest that absolute seizure freedom is the only outcome consistently associated with improvement in quality of life. A meta-analysis showed that the overall weighted pooled-risk difference in favor of the newer antiseizure medications over placebo for seizure-freedom during the limited study periods was only 6% (10).

Adverse effects

Adverse effects are a leading cause of treatment failure with antiseizure medications. They result in early treatment discontinuation in up to 25% of patients and may preclude attainment of fully effective doses and have a negative effect on patient adherence. Furthermore, adverse effects of antiseizure medications are a major source of disability, morbidity, and mortality and a substantial burden on use and costs of health care. Substantial effort has been made to define, quantify, and address their clinical relevance (65; 60).

According to the World Health Organization classification, there are five types of medication-related adverse effects:

Some of the more common or serious adverse effects are shown in Table 6. Serious and life-threatening adverse effects are highlighted by the Food and Drug Administration in boxed warnings of the package inserts.

Table 6. Primary Adverse Reactions of Antiseizure Medications: Sometimes Serious and Rarely Life-Threatening

The table above is based predominantly on information obtained from the summaries of product characteristics (SmPCs3-19) and package inserts (57). It is an assessment of common adverse medication reactions and those of clinical importance. It is not an exhaustive list of all adverse medication reactions; for these, readers should refer to the summaries of product characteristics or package inserts of each antiseizure medication. There is a comprehensive review of the safety profile of the newest antiseizure medications: brivaracetam, eslicarbazepine acetate lacosamide, and perampanel (56).

Type A effects can be ascribed to the known mechanism of action of the medication; they usually arise at the beginning of treatment or after dose escalation and typically abate over time or after dose reduction. The most representative type A effects of antiseizure medications affect the CNS and include drowsiness, fatigue, dizziness, unsteadiness, blurred or double vision, difficulty concentrating, memory problems, irritability, and depression. These are often dose-related and can be anticipated by adopting a slow-titration schedule. It is unclear why some patients develop intolerable neurotoxic side effects at starting doses whereas others can tolerate substantial amounts of the same agent with impunity (16). Genetic factors likely play a role.

Type B effects (idiosyncratic). The most common idiosyncratic reaction to antiseizure medications is skin rash, which can range from a trivial evanescent exanthem to life-threatening severe cutaneous reactions (Stevens-Johnson syndrome and toxic epidermal necrolysis) (04). Maculopapular rashes are noted in 5% to 17% of patients started on carbamazepine, phenytoin, phenobarbital, and lamotrigine. Severe mucocutaneous reactions, such as drug rash with eosinophilia and systemic symptoms (DRESS), Stevens-Johnson syndrome, and toxic epidermal necrolysis, affect 1 to 10 in 10,000 new users of these antiseizure medications. Risk factors for cutaneous reactions include genetic predisposition, pediatric or old age, a previous history of cutaneous reactions with other medications, high initial dose and rapid escalation schedules, concomitant immune system disorders, and specific concurrent medications. Presence of the human leukocyte antigen HLA-B 1502 is strongly correlated with carbamazepine-induced Stevens-Johnson syndrome and toxic epidermal necrolysis in patients with ancestry across broad areas of Asia, and evidence suggests these findings can also apply to phenytoin, lamotrigine, and oxcarbazepine. Among other HLA genotypes that predispose to antiseizure-induced reactions, the antigen HLA-A 3101 has been found to be associated with multiple carbamazepine-induced cutaneous reactions in Chinese, Japanese, and European populations.

Other serious idiosyncratic reactions to antiseizure medications include hepatotoxic effects with valproate, histologically evident as a microvesicular steatosis, in less than 1 in 20,000 patients, but this reaction is of particular concern in children under 3 years of age receiving multiple antiseizure medications (22). Felbamate usage is associated with aplastic anemia and hepatotoxicity in around 1 in 4000 patients. Rare cases of acute angle-closure glaucoma and myopia have been reported with topiramate (24).

Type C effects (long-term complications). Type C effects include chronic reactions related to cumulative medication exposure. They can be insidious because of their slow progressive development. Although some of these effects recede after discontinuation of the medication, others can be irreversible. Long-term complications of antiseizure medications tend to be medication specific. Chronic usage by many of the older agents can lead to dysmorphic changes, such as gum hypertrophy and hirsutism with phenytoin (15) and weight gain and polycystic ovarian syndrome with valproate (49). Weight gain is particularly associated with valproate, gabapentin, pregabalin, vigabatrin, and perampanel. Weight loss is associated with topiramate and zonisamide.

A serious type C effect is bilateral visual field loss induced by vigabatrin, which is irreversible and can worsen with continued treatment. Prevalence ranges from 14% to 92%. Suggested risk factors include male sex, older age, cumulative dose and mean daily dose of vigabatrin, and duration of exposure.

Treatment with topiramate and zonisamide carries a small risk of renal calculi via their shared carbonic anhydrase-inhibiting activity. This property may also account for their association with hypohidrosis, which can be symptomatic for patients living in tropical countries (08).

Long-term antiseizure medication therapy can produce hypocalcemia and decrease biologically active vitamin D levels, leading to reduced bone mineral density and a higher risk of fractures. Both enzyme-inducing and nonenzyme-inducing agents are implicated, and their effects may be additive. A variety of mechanisms have been suggested, the most significant of which is an increased rate of bone turnover. Patients at high risk of osteoporosis, eg, those on polytherapy or with poor mobility, are recommended to take calcium and vitamin D supplements and to have regular bone density measurements (01).

Psychiatric complications. Adverse psychiatric effects are noted in about 15% to 20% of patients with epilepsy who take antiseizure medications. These effects include behavioral or personality changes (irritability, hyperactivity, agitation, and aggressive behavior), depression, and psychosis. Some antiseizure medications, such as phenobarbital, topiramate, and vigabatrin are associated with neuropsychiatric complications and should be used cautiously in patients with a history of mental illness. Behavioral problems such as agitation, aggression, hostility, psychosis, anxiety, and depression have been reported in up to 7% of patients treated with levetiracetam, particularly those with a prior history of psychiatric illness (75). Zonisamide can also produce or exacerbate psychiatric comorbidities, particularly depression (18).

Furthermore, growing evidence suggests that individual vulnerability plays an important part. For example, family and personal psychiatric history, family history of epilepsy, personal history of febrile convulsions, and presence of tonic-atonic seizures are independent predictors for occurrence of psychiatric symptoms during treatment with topiramate. A positive psychiatric history seems to be an important risk factor for development of adverse psychiatric effects with other antiseizure medications (60). The United States Food and Drug Administration analyzed data from 199 placebo-controlled trials of 11 antiseizure medications and found an increase in risk of suicidal behavior or ideation among patients receiving antiseizure medications (0.43%) compared to those receiving placebo (0.22%) (US Food and Drug Administration 2008). There were four completed suicides among the 27,863 patients randomized to receive antiseizure medications compared with none among the 16,029 in the placebo groups. Although the difference was statistically significant, the absolute magnitude of the difference was small considering the sample size analyzed. Given that antiseizure medication therapy was evidently indicated for these patients, the implication of such findings on treatment alternatives is unclear. Whether other factors, such as indication of treatment- and medication-specific effects, may play a role is controversial. A subsequent large-scale community-based study found that the current use of antiseizure medications was not associated with an increased risk of suicide-related events among patients with epilepsy, but it was associated with an increased risk of such events among patients with depression and among those who did not have epilepsy, depression, or bipolar disorder (03). A community-based study of patients with treated epilepsy documented increased risk of suicidal behavior only in those receiving the newer antiseizure medications with a high risk of depression (levetiracetam, tiagabine, topiramate, vigabatrin) (02). These reports have raised renewed concern over adverse psychiatric effects of antiseizure medications in general, and it is advisable that patients receiving antiseizure medications should be monitored for the emergence or worsening of suicidal thoughts or behavior.

Type D effects. Teratogenic and carcinogenic effects are included in the type D category. It is generally accepted that antiseizure medication treatment during the first trimester of pregnancy is associated with a small, but significant, increase in the risk of major congenital malformations (defined as structural abnormalities with surgical, medical, or cosmetic importance). This risk is:

See details in articles on Pregnancy and epilepsy as well as in Fetal anticonvulsant syndrome.

Adverse cardiac effects of antiseizure medications. There is a conspicuous poverty of information and recommendations on the cardiac adverse medication reactions of antiseizure medications and their effect on the electrocardiogram. The situation may now change in view of the increasing number of studies on SUDEP (05) and the attention now given to cardiac function when assessing medication safety during the regulatory process, particularly with respect to ventricular repolarization as reflected in the prolongation of the ECG QT interval. Premarketing investigation of the safety of a new pharmaceutical agent now includes rigorous characterization of its effects on the QT/QTc interval labelled “Thorough QT/QTc Study” (57; 68). Antiseizure medications of sodium channel blockade contribute to ECG changes and probably SUDEP (05).

Table 7. Adverse Cardiac Effects of Antiseizure Medications

Table 7 information presented as cited in the summaries of product characteristics; modified from (57).

Type E effects (adverse medication interactions). Adverse medication interactions are common with epilepsy treatment and are usually clinically relevant because:

Long-term outcome. Long-term outcome studies suggest that 60% to 70% of newly diagnosed patients will become seizure-free with antiseizure medication therapy (40), mostly while taking their first or second antiseizure medication regimen (35; 52; 14). In line with this observation, the International League Against Epilepsy (ILAE) has proposed a consensus definition to define refractory or medication-resistant epilepsy as “a failure of adequate trials of two tolerated and appropriately chosen and used antiseizure medication schedules (whether as monotherapies or in combination) to achieve sustained seizure freedom” (34). It should be noted that the classification of a patient’s epilepsy as medication-resistant at a given point in time does not invariably imply that the patient will never become seizure-free on further antiseizure medication therapy (19; 44). Indeed, although approximately 60% of patients with newly diagnosed epilepsy appeared to have a constant course (either early remission or never seizure free), prognosis in the other 40% was less predictable, with some becoming seizure free after a delay and one in six demonstrating a “remitting-relapsing” course, fluctuating between periods of seizure freedom and recurrence (14). Thus, medication response can be a dynamic process. At least in the short term, prognosis for patients who do not respond to their first few medication regimens should not always be viewed nihilistically. Nonetheless, patients with epilepsy fulfilling the definition should be referred for a comprehensive review of the diagnosis and management, preferably at a specialist center. Workup for epilepsy surgery can be considered at this stage, particularly if a potentially operable structural abnormality, such as mesial temporal sclerosis, has been identified (17). If non-medication treatment options are not appropriate, further pharmacological modifications can still occasionally produce a worthwhile and sustained response.

Medication can be withdrawn slowly without seizure recurrence (at least over the following several years) in around 60% to 70% of adults who have been seizure-free for 2 years or more (47). Patients reporting few seizures before and after starting treatment, those taking a single medication, those seizure-free for many years, and those with no structural abnormality on brain imaging are likely to have the best outcomes. Myoclonic epilepsies carry a high risk of recurrence.

The value of the EEG in predicting relapse is small, particularly in adults (48). A normal EEG does not mean that antiseizure medication withdrawal is safe. Conversely, ictal EEG abnormalities associated with clinical manifestations (eg, jerks or absences) is a definite indicator for continuing proper antiseizure medication treatment.

The decision of whether to withdraw treatment should be made, whenever possible, by the patient, taking into account the likelihood of relapse and its potential psychosocial impact (eg, on driving and employment) balanced against the risk of adverse events from continued treatment (eg, teratogenicity) (57).

Consideration of total withdrawal of antiseizure medications is needed in the following patients:

Discontinuation of antiseizure medications should be extremely slow, in small doses and in long steps of weeks or months. Data in children suggest that a rate of 25% every 10 days to 2 weeks is safe (30). The rate of relapse increases with a faster rate of antiseizure medication discontinuation. Furthermore, with fast discontinuation of antiseizure medications, there is a risk of seizures that are directly related to the withdrawal effects of certain medications (phenobarbital and benzodiazepines). Before antiseizure medication withdrawal, there is a need for a thorough re­evaluation of the patient, including an EEG in children (30). The presence of even minor and infrequent seizures specifies active disease. Conversely, the occurrence of such seizures in the process of medication discontinuation mandates restoration of antiseizure medication.

A significant report presents evidence-based nomograms to facilitate prediction of outcomes following medication withdrawal for the individual patient, including both the risk of relapse and the chance of long-term freedom from seizures (41). Independent predictors of seizure recurrence were epilepsy duration before remission, seizure-free interval before antiseizure medication withdrawal, age at onset of epilepsy, history of febrile seizures, number of seizures before remission, absence of a self-limiting epilepsy syndrome (childhood absence epilepsy, Rolandic epilepsy, or Panayiotopoulos syndrome), developmental delay, and epileptiform abnormality on EEG before withdrawal. Independent predictors of seizures in the last year of follow-up were epilepsy duration before remission, seizure-free interval before antiseizure medication withdrawal, number of antiseizure medications before withdrawal, female sex, family history of epilepsy, number of seizures before remission, focal seizures, and epileptiform abnormality on EEG before withdrawal (41).

Over-medication in terms of the number of antiseizure medications and doses and length of exposure is undesirable but common. Treatment should be reviewed at regular intervals to ensure that patients are not maintained for long periods on medications that are ineffective, poorly tolerated, or not needed and that concordance with prescribed medication is maintained.

Special considerations

Although it would be ideal for a woman contemplating pregnancy to have antiseizure medication treatment withdrawn, for many this would result in recurrence or exacerbation of seizures, which can be dangerous for both mother and baby. If the criteria for discontinuation are met, this should be done prior to conception in a planned pregnancy counseling program. If antiseizure medications cannot be withdrawn completely, treatment should be tapered to a minimally effective dose of, if possible, a single medication. See details in the article on Pregnancy and epilepsy.

Scientific basis

Epileptic seizures result from abnormal, excessive, and synchronous excitation of neuronal populations. Antiseizure medications suppress their occurrence by modifying the bursting properties of neurons and by reducing hypersynchronization, thereby inhibiting the spread of abnormal neuronal firing (64). Although an antiepileptogenic effect has been noted for some antiseizure medications in selected animal models (66), such a property has not been demonstrated in clinical trials (72).

The mechanisms of action of the currently marketed antiseizure medications involve alterations in the balance between neuronal excitation and inhibition (Table 8). Three basic mechanisms are recognized: (1) modulation of voltage-gated ion channels (including sodium, calcium, and potassium channels), (2) enhancement of synaptic inhibitory neurotransmission, and (3) attenuation of brain excitation (Table 8). Many antiseizure medications have multiple modes of action, whereas the mechanisms for others are still unclear.

Table 8. Main Mode of Action of Antiseizure Medications

Table 8 information based on data from summaries of product characteristics and product inserts; modified from (57).

Voltage-gated ion channels regulate the electrical behavior of neurons as well as the release of neurotransmitters. Medications such as phenytoin, carbamazepine, and lamotrigine act primarily by limiting sustained, repetitive firing via blockade of voltage-gated sodium channels. This property is shared by some of the newer medications, such as oxcarbazepine, zonisamide, and eslicarbazepine. Gabapentin and pregabalin exert their antiseizure and other pharmacological effects by binding to the alpha2-delta subunit of the calcium channel.

Several antiseizure medications act primarily by enhancing GABAergic inhibition. The barbiturates and benzodiazepines modulate GABAA receptors. Vigabatrin and tiagabine inhibit the enzymatic degradation and reuptake of GABA, respectively.

Levetiracetam (binding to synaptic vesicle protein SV2A), lacosamide (selectively enhancing slow inactivation voltage-gated sodium channels), and perampanel (selective, noncompetitive, AMPA receptor antagonist) have novel modes of action.