Once the exact type of seizure has been determined, the physician should initiate monotherapy (single-drug therapy) with the drug that has shown the best combination of high efficacy and low toxicity in comparative studies of drugs for the patient’s seizure type (Table 10-1). If a patient has more than one type of seizure, therapy should begin with the drug of choice for the combination of seizure types present (Table 10-1). Reasons for selection of specific drugs as drugs of choice for specific seizure types are given in section IV.

The single-drug approach for initial treatment is preferred because monotherapy with an appropriately selected drug pushed to the adequate dosing rate controls seizures in 60% to 90% of patients, and polytherapy (therapy with more than one drug) exposes the patient to several unnecessary risks. Factors that appear to be associated with failure of monotherapy include persistent noncompliance, drug allergy, large or progressive brain lesions, partial seizures, more than one type of seizure, neuro-psychiatric handicaps, and high pretreatment seizure frequency.

The risks of unnecessary polytherapy are many. Chronic toxicity is associated with the use of any antiepileptic drug. Polytherapy may include barbiturates, which are associated with
high risk for cognitive and behavioral toxicity. Other risks of unnecessary polytherapy include drug allergy, drug interactions, exacerbation of seizures, noncompliance, costs, and the inability to evaluate the effectiveness of individual antiepileptic drugs.

The first drug tried for a seizure disorder is usually among the least-toxic drugs available, and the physician must be certain that the maximal possible therapeutic effect has been obtained from the first drug before adding other drugs. Therapy usually begins with a so-called average dosage of antiepileptic drug. If the seizures are controlled with this average dosage and no serious side effects appear, no further changes are necessary. If the seizures are not controlled with this dosage and no serious drug toxicity occurs, the dosage of the drug should be systematically increased until the seizures are controlled or until side effects preclude further dosage increase.

The drug plasma concentration should be determined if a patient’s seizures are not controlled by an average or high drug dosage. Many correctable causes of lower than expected drug plasma concentration exist, including inadequate dosing rate, noncompliance, poor absorption, drug interactions, generic drug substitutions, pregnancy, and patient error. To substitute or to add a more toxic drug because the patient has a low plasma concentration of the first drug is a serious error. A drug cannot be said to be ineffective until it is documented that the seizures are not controlled with a high therapeutic plasma concentration of the drug, unless drug toxicity precludes reaching such concentrations.

The therapeutic range of drug plasma concentration represents values applicable to average patients. Some patients require higher drug plasma concentrations than the therapeutic range
for good seizure control. If a patient has a high therapeutic plasma concentration of a nontoxic drug, poor seizure control, and no drug side effects, the best approach usually is to increase the dosage of the first drug rather than to add a more-toxic second drug.

If the first drug is pushed to its maximal tolerated dosage and seizures still are not controlled, a second antiepileptic drug should be added. In general, adding the second drug and continuing administration of the first drug (at least temporarily) is best because (a) the first drug will provide protection while the plasma concentration of the second drug is being built up, (b) discontinuing the first drug may result in withdrawal activation of the seizure disorder, and (c) evidence exists that two antiepileptic drugs in combination may control seizures in some patients when either drug alone does not.

When a therapeutic plasma concentration of the second drug is obtained, the physician should consider tapering the patient off the first drug because of the many hazards of prolonged polytherapy. The decision to taper the patient off the first drug must be individualized and should take into consideration the antiepileptic effect of the first drug when it is given alone, the side effects of the first drug, and the psychosocial consequences to the patient of having a seizure if discontinuation of the first drug results in loss of complete seizure control. The adverse effects of antiepileptic drugs on behavior and cognition argue that the physician should attempt to minimize the number of these drugs given to children. In adults, the hazards of polytherapy must be weighed against the risk of loss of job or driver’s license if discontinuation of the first drug results in a recurrence of seizures. If discontinuation of the first drug is elected, it should be done slowly.

A third drug should not be added until it is documented that seizures cannot be controlled with maximal tolerated doses of the first two drugs tried. Adding a third drug (at least temporarily) is usually better than substituting a third drug for the first or second drug, for reasons similar to those cited earlier for adding, rather than substituting, a second drug. After a therapeutic plasma concentration of the third drug is reached, the physician may elect to discontinue one of the first two drugs, by using the guidelines just outlined.

Simple partial (focal); complex partial (psychomotor, temporal lobe); and tonic-clonic (grand mal) seizures are the most common types of seizure disorders and occur at all ages. Note that a tonic-clonic seizure may be focal (partial seizures secondarily generalized) or generalized (primarily generalized) in onset (see Chapter 2). Carbamazepine (Tegretol), phenobarbital, phenytoin (Dilantin), primidone (Mysoline), and valproic acid (Depakote)
are the drugs that have been used as initial therapy for simple partial, complex partial, and partial seizures secondarily generalized. These five drugs have been compared in adults in two large Veterans Administration Cooperative studies. Carbamazepine and phenytoin had the best combined efficacy and safety results. Primidone was inferior to the other four drugs for all seizure types because of a significantly higher incidence of intolerable toxicity. More recently, oxcarbazepine (Trileptal) has been compared with carbamazepine, phenytoin, and valproic acid. All drugs had similar efficacy, but oxcarbazepine has the fewest side effects.

Carbamazepine, oxcarbazepine, and phenytoin are the three drugs of choice for partial seizures and for partial seizures secondarily generalized in adults, based on the following: (a) carbamazepine, oxcarbazepine, and phenytoin have fewer side effects than phenobarbital or valproic acid, regardless of seizure type; (b) carbamazepine and phenytoin are more effective than phenobarbital or valproic acid for complex partial seizures; (c) no statistically significant differences were seen in efficacy when carbamazepine, oxcarbazepine, and phenytoin were compared; and (d) monotherapy with carbamazepine, oxcarbazepine, or phenytoin produces a satisfactory long-term result in approximately 60% to 80% of patients. Gabapentin, lamotrigine, and topiramate are not approved by the U.S. Food and Drug Administration (FDA) for initial therapy of partial seizures. Lamotrigine is approved for conversion to monotherapy for partial seizures.

In children, comparative studies have demonstrated that carbamazepine, phenytoin, and valproic acid are equally efficacious. Phenobarbital and primidone also are efficacious, but side effects, such as irritability, hyperactivity, and lethargy, limit these drugs to second-line therapy. In general, carbamazepine is preferred over phenytoin because of the erratic absorption in children, resulting in fluctuating blood levels and cosmetic side effects such as gingival hypertrophy and hirsutism. Because it is sometimes difficult to distinguish between partial seizures and primarily generalized seizures in infants, valproate may be an appropriate first choice in some children.

Gabapentin, lamotrigine, levetiracetam, oxcarbazepine, topiramate, tiagabine, and zonisamide all have been tested as adjunctive therapy for refractory partial seizures in children, with promising results.

No definitive trials have been published establishing the drug of second choice for patients whose first-choice drug (carbamazepine, oxcarbazepine, or phenytoin) has failed a trial. Probably the most common practice is to add or substitute another first-choice drug based on the results of initial therapy studies. This practice has been questioned because (a) carbamazepine and phenytoin have complex drug interactions when taken together, and (b) all three drugs have the same mechanism of action.

Three older drugs (phenobarbital, primidone, valproic acid) and seven new drugs (gabapentin, lamotrigine, levetiracetam, oxcarbazepine, tiagabine, topiramate, zonisamide) have been used as alternative drugs in patients failing to respond to carbamazepine, oxcarbazepine, or phenytoin. The advantages and disadvantages of these drugs are summarized in Table 10-2 and the review by Cramer et al. (12). Phenobarbital and primidone have fallen from favor because of the high incidence of cognitive/behavioral side effects and drug interactions. Although used by many experts, valproic acid is not more effective than newer agents in adults and shows more problems with serious toxicity, nuisance toxicity, and drug interactions.

Few comparative studies of the new antiepileptic drugs are available. The information here and in Table 10-2 is based on separate studies in adults using varying methods. The results of these studies may not be strictly comparable and may not apply to children.

Gabapentin and levetiracetam have no serious drug toxicity, little nuisance toxicity, and no drug interactions. Lamotrigine is well tolerated by many patients and has few drug interactions with non-antiepileptic drugs. Topiramate has the highest reported responder rate and few drug interactions. Tiagabine has the lowest responder rate, several side effects, multiple drug interactions, and inconvenient administration. Topiramate and zonisamide can cause kidney stones.

The opinion of many experts is that the efficacy-to-toxicity ratios for valproic acid may be more favorable in children than in adults.

For primarily generalized tonic-clonic seizures, carbamazepine, phenytoin, and valproic acid are all effective as initial therapy. Topiramate is effective as adjunctive therapy for primarily generalized tonic-clonic seizures. Valproic acid has efficacy against absence and myoclonic seizures (sometimes associated with primarily generalized tonic-clonic seizures), whereas carbamazepine and phenytoin do not. Absence seizures may worsen in some patients taking carbamazepine or phenobarbital. Valproic acid is the drug of first choice for persons with primarily generalized tonic-clonic seizures with absence seizures, myoclonic seizures, or both. Lamotrigine, gabapentin, levetiracetam, oxcarbazepine, tiagabine, and zonisamide have shown promise against primarily generalized tonic-clonic seizures in a limited number of studies.

Ethosuximide, valproic acid, and clonazepam are the three drugs used to treat absence seizures (Table 10-1) and are equally effective for that purpose. Ethosuximide has the fewest side effects and is the first-choice drug for uncomplicated absence seizures. For patients with tonic-clonic seizures, myoclonic seizures, or both, in addition to absence seizures, valproic acid is the drug of first choice (ethosuximide has no efficacy for tonic-clonic or myoclonic seizures). Lamotrigine has been reported to be effective for absence and primarily generalized tonic-clonic seizures (it is not yet approved by the FDA for these indications) and is used by some experts for the combination of absence and tonic-clonic seizures. Clonazepam is not a favored therapy because of side effects (drowsiness, hyperactivity); tolerance to antiabsence effect; and equivocal efficacy against tonic-clonic seizures.