Pocket Drug Guide

Archive for the ‘Anti-Epileptic Drugs’ Category

Phenytoin

Phenytoin has been used for over 50 years. Thus, the characteristics and side-effects have been well elucidated. Its primary use is for treatment of partial epilepsy, including simple and complex partial seizures, as well as generalized tonic-clonic seizures, whether of primary or partial onset. Phenytoin can be initiated with a ‘loading dose’ of 13-20mg/kg, or with a starting dose of 3-5 mg/kg/day. Phenytoin was one of the first anti-epileptic drugs that became associated with a ‘therapeutic range’ (10-20 mg/1 in most laboratories), which represents serum concentrations most likely to produce a therapeutic effect without substantial dose-related side-effects. The phenytoin dose should be selected using therapeutic monitoring rather than a predetermined dose, such as 300 mg/day. As the serum level increases, side-effects such as lethargy, ataxia, dysarthria, fatigue, diplopia, abnormal movements, mental confusion and cognitive changes may occur. This is particularly true at concentrations above 20 mg/1. However, the therapeutic range represents an average. A number of patients may remain seizure-free at serum concentrations below the range, or may tolerate levels substantially above the range. However, it is very important to understand the properties of phenytoin metabolism. As serum concentrations rise, particularly to levels above 15 mg/1, the metabolism of phenytoin slows substantially. This is called ‘zero order kinetics’. At levels below 15 mg/1, doubling the dose will lead to a doubling of serum concentration, and the half-life is 24 h. As metabolism slows at higher concentrations, even a 50-mg change in dose can double the serum concentration, and the half-life increases to 48-70 h. Since the half-life is so prolonged, a steady state after dose adjustments may not occur for weeks, with serum levels slowly rising over this time. This can easily lead to serious phenytoin toxicity. Hospitalization due to phenytoin toxicity under these circumstances is not uncommon. Dosage adjustments must be made with great care, and levels should be repeated a month after any adjustment. In addition to dose-related toxicities, patients receiving phenytoin may experience side-effects that are relatively independent of dose. These include gingival hyperplasia, acne and hirsutism. Visits to the dentist every 6-12 months, accompanied by daily flossing, can help prevent gingival hypertrophy. There is now fairly substantial evidence that long-term phenytoin use can lead to reduced bone density and risk of fracture. Obviously, this is a major issue for patients who fall with their seizures. All patients receiving phenytoin should receive supplemental calcium and vitamin D, and should be screened with bone densitometry. Allergic rash may occur. Although idiosyncratic side-effects are not common, they can occur, and patients should be advised of this possibility. These include Stevens-Johnson syndrome, aplastic anaemia, hepatic failure and a lupus-like syndrome. Monitoring of blood counts, liver function tests and electrolytes are warranted for the first 6-12 months of therapy. Drug interactions are relatively common, both with other epilepsy drugs and with drugs taken for other conditions. Interactions with other anti-epileptic drugs are listed in Because of phenytoin’s relatively long half-life, it can be administered only once or twice a day.

Carbamazepine

Carbamazepine has long been considered a first-line agent for simple partial, complex partial and generalized tonic-clonic seizures. Recently, many studies have compared the newer anti-epileptic drugs to carbamazepine as the ‘gold standard’ in patients with newly diagnosed epilepsy, and to date, none has been found to be more effective.

Table Pharmacokinetic impact of old anti-epileptic drugs (anti-epileptic drugs) on new anti-epileptic drugs. -I, levels decrease (faster clearance); T, levels increase (slower clearance)

	Gabapentin/ pregabalin	Lamotrigine	Topiramate	Tiagabine	Levetiracetam	Zonisamide	Oxcarbazepine (MHD)
Phenytoin/carbamazepine/ phenobarbital/primidone	None	↓	↓	↓	None	↓	↓
Valproate	None	↑	None	None	None	None	Slight ↓

Table Pharmacokinetic impact of new anti-epileptic drugs (anti-epileptic drugs) on old anti-epileptic drugs. -I, levels decrease (faster clearance); T, levels increase (slower clearance)

	Phenytoin	Carbamazepine	Valproate	Phenobarbitat	Primidone
Gabapentin/tiagabine/ zonisamide/levetiracetam/	None	None	None	None	None
pregabalin
Lamotrigine	None	None	I 25%	None	None
Topiramate	May t	None	None	None	None
Oxcarbazepine	May t	None	None	Slight t	None

However, lamotrigine, topiramate and oxcarbazepine all caused fewer dropouts due to side-effects than carbamazepine. Of note, it is considered ‘narrow spectrum’, and may actually worsen generalized epilepsies, particularly those associated with absence or myoclonus. Initiation must be done with titration, to avoid appearance of dose-related side-effects. Dose selection should be done using clinical response and serum levels. Typically, a level of 8-12mgA will provide the best response, and few patients will tolerate serum concentrations above 15mg/l. Dose-related adverse effects include ataxia, drowsiness, vertigo, difficulty concentrating, diplopia and blurred vision. Using sustained-release formulations of carbamazepine will reduce some of these side-effects by reducing peak serum concentrations. The other advantage is the ability to use twice-a-day dosing, rather than the three- to four-times-a-day dosing that would otherwise be necessary due to the short half-life of carbamazepine. Gastrointestinal symptoms such as nausea, vomiting, diarrhoea and constipation may also be seen. The other side-effects of carbamazepine are not necessarily linked to dose. These include more common, but less serious, effects such as leucopenia, hyponatraemia and rash as well as life-threatening but extremely rare idiosyncratic reactions such as Stevens-Johnson syndrome, aplastic anaemia and hepatic failure. Mild leucopenia can be disregarded. Rarely, more profound leucopenia (with leucocyte counts <2.5xl0⁹) may necessitate discontinuation of therapy. Hyponatraemia does not necessitate discontinuation in all patients, as it can often be managed with water restriction. However, in some patients water restriction does not work, or too severely impacts lifestyle, and in these patients an alternative anti-epileptic drug should be sought. Because of the potential for all of the above issues, monitoring of liver function tests, white blood counts and electrolytes are necessary during the first 6-12 months of therapy, and yearly thereafter. The potential for carbamazepine to produce decreased bone density is under study. Of note, chronic carbamazepine use has been associated with some reduction in serum testosterone levels and increases in cholesterol levels. The clinical implications of these changes are under study. As with phenytoin, carbamazepine is associated with many drug interactions affecting anti-epileptic drugs and other drugs. Because many drugs can inhibit or induce the metabolism of carbamazepine, and cause acute toxicity or reduced effect, it is a good idea to tell patients to inform their neurologist when starting any prescription drug. Carbamazepine, even in the extended-release formulations, is substantially less expensive than the newer anti-epileptic drugs.

Phenobarbital

Phenobarbital is the ‘grandfather of anti-epileptic drugs’. It has been available for over 100 years. It is effective for most seizure types and the fact that an intravenous (i.v.) formulation is available means it is often used for treatment of status epilepticus. In the modern era, it is rarely used as first-line therapy, as studies have demonstrated that it causes more dose-related side-effects, particularly sedation, than other options. Also, once started it is very difficult to withdraw without causing seizure exacerbation. Abrupt withdrawal is not recommended, and even slow withdrawal can lead to problems. The initial dose is 30-50 mg, which is best administered at bedtime. Titration to optimal dose can be achieved over several weeks. Optimal effect is usually achieved at serum concentrations of 15-45 mg/1. Other dose-related side-effects include irritability, difficulty concentrating, memory loss, sedation, dysarthria and ataxia. Other reported adverse reactions include hyperactivity, mostly in children, and depression. As with phenytoin and carbamazepine, hypersensitivity syndrome, rash, hepatic failure and aplastic anaemia may occur. Since many patients who are currently on phenobarbital have been on it long term, an understanding of long-term side-effects is important. Some unique side-effects occur with long-term phenobarbital use, affecting the skin and connective tissue. These may include contractures, frozen shoulder and general pain. Drug interactions may occur with phenobarbital . It is the least expensive anti-epileptic drug, costing pennies a day, and its long half-life permits once-a-day dosing, which is important for potentially non-compliant patients.

Primidone

Primidone is no longer considered a first-line drug, and its use has diminished substantially. It is metabolized to a number of active metabolites, including phenobarbital and phenylethylmalonamide, although the parent compound is also active. When obtaining serum levels, therapeutic effect will be associated with both the parent primidone levels and phenobarbital levels. In the presence of enzyme-inducing drugs, primidone levels will go down, and phenobarbital levels will rise. Primidone causes all the problems seen with phenobarbital use, described above.

Valproate

Valproate has been available since the 1970s, but remains the first-line drug for many epilepsy syndromes, including juvenile myoclonic epilepsy and syndromes associated with absence seizures. Valproate may be the only effective agent for some patients. A recent large randomized open-label trial demonstrated superior efficacy compared with lamotrigine, and superior tolerability compared with topiramate, in adults and children with idiopathic generalized or unclassified epilepsy. Seizure control was maintained in the majority for 5 years.

In some countries, valproate is also used as first-line therapy for partial seizures. In a head-to-head study, valproate was as effective for generalized tonic-clonic convulsions as carbamazepine in patients with partial epilepsy, but was found to be less effective for complex partial seizures. However, in other studies, the efficacy was equivalent in all seizure types, when valproate was compared with phenytoin, carbamazepine or oxcarbazepine. Valproate is available in a variety of formulations, including sustained release, sprinkles and i.v.. With the availability of i.v., valproate has become more popular for treatment of status epilepticus. Typical initiation in outpatients is at a dose of 10—15mg/ kg/day, with subsequent increases of 5-10mg/kg/week as needed to control seizures. Serum levels between 50 and 100 mg/1 are typically therapeutic and well tolerated, but higher concentrations may be necessary in some patients.

Skin rashes are less common with valproate than with carbamazepine, phenytoin and lamotrigine, and therefore valproate is a reasonable choice for patients with a history of hypersensitivity. The main side-effects are Gastrointestinal upset, nausea, vomiting, tremor, weight gain and hair loss. Tremor and weight gain are dose related. Gastrointestinal upset can sometimes be avoided by using alternative formulations, such as sustained release or sprinkles. Elevated ammonia levels can be seen commonly, affecting 20-50% of patients. In some patients this is well tolerated and not problematic, while in others elevated ammonia can be associated with encephalopathy, triphasic waves and even coma. It is not recommended to monitor ammonia in asymptomatic patients. However, if patients demonstrate encephalopathy, particularly in the presence of asteryxis and/or delta slowing on the electroencephalogram, it is reasonable to check for hyperammonaemia. Carnitine supplementation has been recommended to improve valproate-induced hyperammonaemia. Idiosyncratic side-effects include rare fatal hepatotoxicity. The incidence is 1 in 20 000 in adults on monotherapy, but much more common in polytherapy, in children under 10 years old, and even more so in children under 2 years old. Children who have seizures resulting from metabolic disorders may be at extremely high risk, and as a rule should not be treated with valproate. When children on polytherapy under the age of 2 are treated, the reported incidence is as high as 1 in 600-800. The greatest risk is in the first 6 months of use, and clinical as well as blood monitoring is indicated during this time. Pancreatitis may also occur, which may have a frequency of up to 1 in 3000. The risk of pancreatitis does not diminish over time. Aplastic anaemia is rare, but thrombocytopenia and altered bleeding time are common. These are usually not clinically worrisome, and low platelet counts can usually be tolerated as they do not portend more significant blood dyscrasias. Some case reports note increased risk of bleeding during surgery in patients on valproate, whereas many others note no increase in bleeding. Recently, a number of studies have reported adverse fetal outcomes in offspring of women receiving valproate. Spina bifida is seen in up to 2% of children exposed to valproate before birth. This makes valproate a poor choice for women contemplating pregnancy. In some studies, valproate has also been associated with development of polycystic ovarian syndrome. This appears to be more prevalent when valproate is initiated in women under 40. Reports of menstrual cycle irregularities are common.

One property of valproate that must be kept in mind is that it is a hepatic metabolic inhibitor. This has a number of consequences, including a number of interactions with other anti-epileptic drugs as well as drugs of different classes. In addition, intrinsic substances such as oestrogen may be inhibited. Valproate is highly protein bound. It may displace phenytoin from binding sites, causing emergence of phenytoin toxicity in the absence of an increased plasma level.

Ethosuximide

Ethosuximide has a very narrow therapeutic indication, with use limited to patients with absence seizures. In most cases, it should be used as the sole agent only in patients who experience this seizure type in isolation, a condition seen primarily in childhood. Occasionally, in patients with primary generalized epilepsy with seizure types other than absence, addition of ethosuximide as an adjunctive medication may improve seizure control. Ethosuximide can be started at 500 mg/day, and titrated as tolerated, with weekly increments. Serum concentrations of 40-100 mg/1 are usually optimal. The most common side-effects noted with ethosuximide use include nausea and abdominal discomfort, drowsiness, anorexia and headache. In rare cases, behavioural changes may be seen, including psychosis. Blood dyscrasias have been reported. Drug-drug interactions are minimal.

Benzodiazepines

Available benzodiazepines for chronic use include clonazepam, clorazepate and clobezam. Most benzodiazepines are not good choices for long-term therapy and should not be used as first-line agents. Patients may develop tolerance to the therapeutic effects of many of the benzodiazepines. Seizures may initially be decreased, but as tolerance develops over time, seizures may recur, necessitating dosage increases to regain control. Among the available benzodiazepines, clobezam has been touted as having less propensity for tolerance development. The most common dose-related side-effects of this drug class include drowsiness, ataxia and behavioural problems. Clonazepam is particularly useful for myoclonus. Clorazepate and clobezam may be used for both generalized and partial epilepsies. They tend to be used in patients with drug-resistant epilepsy. Starting dose for clonazepam is 0.5-1 mg twice daily, increasing as needed. Chlorazepate can be initiated at 7.5 mg twice daily or thrice daily. Clobezam can be started at 5mg twice daily and increased by 5-10 mg every 1-2 weeks to achieve the best seizure control without development of excessive sleepiness. Abrupt withdrawal from chronic benzodiazepines may precipitate status epilepticus. Even slow withdrawal may exacerbate seizures. Acute benzodiazepines such as diazepam, lorazepam and midazolam may be used as intermittent ‘rescue therapy’ in patients who experience seizure clusters or status epilepticus.

Drugs are listed in order of approval in the USA, from oldest to newest.

Felbamate

Felbamate is a broad-spectrum anti-epileptic drug. It is not considered to be a first-line drug because of its potential for serious idiosyncratic side-effects, including potentially fatal aplastic anaemia (incidence of 1 in 3000) and hepatic failure (incidence of 1 in 10000). Recent analyses indicate that aplastic anaemia and liver failure occur almost exclusively within the first year of therapy. During this period, safety monitoring consisting of liver function tests and blood counts is recommended with a frequency of up to twice monthly, despite lack of evidence that early detection of changes will prevent serious health problems, even if the drug is discontinued. Yet, felbamate may still be an important drug in the armamentarium for refractory patients, because it may control seizures when other drugs fail, and tends to be alerting rather than sedating. It has been found to be particularly effective in patients with Lennox-Gastaut syndrome. Felbamate should be started at 300-600 mg twice daily, and increased as necessary over several weeks, up to 3600 mg, or higher in some cases. Drug levels maybe useful, and serum concentrations of 30-80mg/1 are recommended. Three-times-a-day dosing may improve tolerability. Dose-related side-effects include insomnia, decreased appetite and weight loss, ataxia, Gastrointestinal upset and headache. Other serious idiosyncratic adverse events include rash and Stevens-Johnson syndrome. Felbamate has a complex metabolism and elimination, by both the hepatic and renal routes, and inhibits the metabolism of some drugs while inducing others. Adjustments in these medications may be necessary.

Lamotrigine

Lamotrigine is a broad-spectrum anti-epileptic drug that has been widely accepted as a first-line drug for both partial and generalized epilepsy syndromes. Clinical trials have supported its use in partial seizures, primary generalized tonic-clonic seizures, absence seizures and seizures associated with Lennox-Gastaut syndrome. There is some controversy regarding the use of lamotrigine in juvenile myoclonic epilepsy. It is commonly used, but may worsen myoclonus in some patients. Lamotrigine has been the subject of a number of head-to-head studies in newly diagnosed partial and generalized epilepsy. Lamotrigine was equally effective but better tolerated than phenytoin and carbamazepine in several somewhat under-powered studies. Several studies indicated that lamotrigine is better tolerated than carbamazepine in the elderly. A recent very large, randomized open-label study indicated that lamotrigine might be the drug of choice in patients with partial epilepsy, as it was equally effective but better tolerated, but in a companion study it was substantially less effective than valproate in patients with generalized or unclassified epilepsy. Some studies have advocated combining lamotrigine with valproate in refractory patients, to achieve maximum efficacy through a favourable pharmacodynamic interaction. However, side-effects are also enhanced when the drugs are combined. Initiation of lamotrigine is somewhat more complex than for other drugs. Slow titration is mandatory to reduce occurrence of serious rash. In monotherapy, initiation with 25mg/day for 2 weeks, then 50mg/day for another 2 weeks, followed by increases of 25-50 mg/week is appropriate. If the patient is receiving valproate at the time of initiation, starting doses and doses during titration should be cut in half, whereas if they are receiving enzyme-inducing anti-epileptic drugs such as phenytoin, carbamazepine or phenobarbital, doses can be doubled. Doses can then be increased as necessary. Typical doses for patients with newly diagnosed epilepsy are 100-200 mg/day. In refractory patients, doses may be as high as 500-1000 mg/day. There is a great deal of variability in serum concentrations that produce optimal effects. Some patients may do well at serum concentrations of 2mg/l, while others may need levels of up to 20 mg/1. Common dose-related side-effects of lamotrigine include mild tremor, double vision, headache and insomnia. Rash, including Stevens-Johnson syndrome and toxic epidermal necrolysis, can occur. Risk is increased for children under 16, and is also more common with concomitant valproate use, and in patients who have experienced rash on other anti-epileptic drugs. Rash almost always occurs within the first 8 weeks of therapy. Other hypersensitivity reactions, although much rarer, may be seen, including lymphadenopathy, fever, hepatic or renal failure, disseminated intravascular coagulation and arthritis. Discontinuation of lamotrigine is recommended in the presence of rash or other indication of hypersensitivity, and it is very important to tell patients to call immediately with such symptoms. Rapid discontinuation can prevent a more severe, potentially life-threatening reaction. In the absence of any clinical symptoms of hypersensitivity, it is unclear that routine monitoring of liver function tests, blood counts or electrolytes is useful. Lamotrigine does not alter the metabolism of other drugs that are given concomitantly. However, other drugs may impact on the metabolism of lamotrigine. One important and common drug interaction that bears noting is that with oral contraceptives, which double the clearance and halve the half-life of lamotrigine. Women on lamotrigine should be told to notify their doctor of any changes in oral contraceptive pill use, something they might not otherwise think to do.

Gabapentin

Gabapentin is a narrow-spectrum drug. It is useful only in patients with partial epilepsy, that is those with simple or complex partial seizures, or partial-onset generalized tonic-clonic convulsions. Gabapentin is felt to be more useful in newly diagnosed patients with mild epilepsy and in the elderly, who benefit from the relatively mild side-effect profile and lack of drug interactions, but less useful in patients with refractory epilepsy. Although gabapentin can be initiated at a therapeutic dose of 1200-2400 mg/day, it appears to be better tolerated when titrated. Typically, 300 mg two times a day is well tolerated as a starting dose, but lower starting doses are needed in some patients. Clinical trials have only tested efficacy to 2400 mg/day. However, in clinical experience, higher doses can be useful. However, gabapentin displays dose-dependent, saturable absorption. At higher doses, less may be absorbed across the gut, leading to diminishing returns. The amount of absorption varies from person to person. Determining if serum levels are rising after dose increments can help ascertain whether increasing doses are futile. Gabapentin is not bound to plasma proteins, is eliminated renally and does not interfere with the metabolism of other medications, including anti-epileptic drugs or psychotropic agents. This makes it an ideal drug for the elderly and patients with chronic illness, who are likely to be taking other drugs. Common dose-related side-effects include somnolence, dizziness, ataxia and fatigue. Weight gain is also seen, and appears to be dose related. Peripheral oedema may occur in some patients. Occurrence of myoclonus has been reported, as well as occasional behavioural disturbance in children. There are no reports of serious idiosyncratic side-effects. Therefore, routine monitoring of liver function tests, blood counts and electrolytes is probably not warranted.

Topiramate

Topiramate is another anti-epileptic drug that is felt to be broad spectrum. It has been studied and found effective in patients with partial-onset or primary generalized tonic-clonic seizures, and in patients with seizures associated with Lennox-Gastaut syndrome. Topiramate is felt to be first-line therapy in all these conditions. One head-to-head study in newly diagnosed patients found topiramate to be equal in efficacy with valproate in patients with mostly generalized seizures, and with carbamazepine in patients with mostly partial seizures, although the study was under-powered and has been criticized methodologically. A recent open-label randomized study in newly diagnosed adults and children indicated that topiramate was as efficacious as any other drug for partial and generalized seizures, but was less well tolerated. However, the open-label nature of the study could have produced some bias. Topiramate should be started at a low dose and slowly titrated for best tolerability, although there are no safety concerns related to starting more rapidly. It is best tolerated when initiated at 25 mg/day and increased by 25 mg/week. Typical doses necessary for newly diagnosed patients are in the range of 100-200 mg, while, as is the case for most drugs, refractory patients may require much higher doses. Up to 1000 mg has been tested in clinical trials, although few can tolerate such high doses. As with lamotrigine, serum levels needed to achieve control are variable, from 2 to 20 mg/1. Monitoring serum levels may be useful. Topiramate is well absorbed and has minimal protein binding. Topiramate is partially metabolized by the liver and approximately 60% is excreted unchanged in the urine. The more common dose-related adverse events include somnolence, paraesthesias (especially of fingertips), fatigue, taste perversion, weight loss and dizziness. One of the side-effects that is relatively specific to topiramate is psychomotor slowing, which particularly affects speech. Patients may complain of word-finding difficulty or slowing of speech. Academic performance can be affected. This side-effect is significant in some patients, while others may escape it entirely, even at high doses. Potential, more serious side-effects that occur infrequently include nephrolithiasis, open-angle glaucoma, causing transient and reversible visual loss, and hypohidrosis in children. Rarely, the hypohidrosis has caused heat stroke with serious consequences. Children receiving topiramate have also rarely developed clinically significant metabolic acidosis. Milder forms are common. Topiramate has rarely been associated with hepatic failure, and this seems to happen more commonly when topiramate is combined with valproate. Topiramate does not impact the metabolism of most concomitant drugs, but it does raise phenytoin levels when added to patients with baseline higher phenytoin levels (e.g. above 15 mg/1), which can potentially cause phenytoin toxicity. Therefore, phenytoin levels should be monitored. In addition, the classic hepatic enzyme-inducing antiepileptic drugs, such as phenytoin and carbamazepine, will increase the metabolism of topiramate, and higher doses may be required.

Oxcarbazepine

Oxcarbazepine, an analogue of carbamazepine, is a narrow-spectrum drug that is considered to be a first-line therapy for the treatment of partial seizures. Although it is similar to carbamazepine, it is effective in some patients for whom carbamazepine has failed and is believed to have additional mechanisms of action. Oxcarbazepine is actually a prodrug for a mono-hydroxylated form, to which it is rapidly converted after oral administration. Several studies have been performed in patients with newly diagnosed epilepsy, comparing oxcarbazepine with older anti-epileptic drugs. In all of these studies, oxcarbazepine was equally efficacious. It was better tolerated than phenytoin and carbamazepine, and as well tolerated as valproate. Oxcarbazepine has been noted to exacerbate myoclonic seizures as well as absence.

Oxcarbazepine should be initiated with titration to avoid side-effects. However, in one inpatient study, 2400 mg/day was started, and caused few dropouts, thus rapid initiation in an emergency is possible. In outpatients, initiation of 300 mg twice daily and titration of 600mg/week is usually well tolerated. Effective doses range from 900 to 2400mg, but higher doses are not well tolerated in combination with other drugs, particularly those with similar side-effect profiles. If side-effects develop as the dose is being increased, spacing out dosing to three times daily, or even four times daily sometimes permits achievement of higher, more efficacious doses. Plasma concentrations of up to 45 mg/1 have been well tolerated. Frequent dose-related adverse events include somnolence, dizziness, headache, ataxia, nausea and vomiting, diplopia, blurred vision, vertigo and tremor. One problematic potential adverse event is hyponatraemia. In a recent study, hyponatraemia was substantially more common for oxcarbazepine (29.9%) than carbamazepine (35.5%), and serum sodium levels of <128mEqA occurred in 12.4%. Risk factors for hyponatraemia include older age and diuretic use. Hypersensitivity syndromes, including rash and Stevens-Johnson syndrome are rare consequences of oxcarbazepine use. Oxcarbazepine does not have the strong enzyme-inducing properties of carbamazepine, and also does not induce its own metabolism. Oxcarbazepine also acts as an inhibitor of phenytoin metabolism, and its own metabolism is induced by the classic inducing anti-epileptic drugs such as phenytoin and carbamazepine.

Tiagabine

Tiagabine is a drug that is considered second-line therapy, and is effective for partial seizures only. Like other drugs that work via y-aminobutyric acid enhancement, tiagabine may exacerbate absence and myoclonic seizures. Tiagabine was compared with carbamazepine in a study of newly diagnosed patients with partial seizures, and was found to be less efficacious. However, it was fairly effective as add-on therapy in patients with refractory epilepsy. Nonetheless, its use has waned in recent years. Of note, some reports of seizures have surfaced when tiagabine has been used off-label for psychiatric indications. Tiagabine is best tolerated when titrated. Initiation of 4 mg once or twice daily is recommended, with dose increments of 4mg/week. A three-four-times-a-day regimen is recommended, due to the short half-life. Doses of up to 64 mg have been used in add-on trials. Patients receiving enzyme inducers will need higher doses than patients taking non-inducing anti-epileptic drugs. Serum levels have not been very clinically useful, due to the short half-life, which results in wide fluctuations in levels from peak to trough. Tiagabine is 96% protein bound, but no protein-binding interactions have been identified clinically. Common dose-related side-effects include tiredness, nervousness, dizziness, headache, tremor and abnormal thinking. An unusual side-effect has been identified, consisting of a stuporous state accompanied by a slow wave or spike-wave pattern on electroencephalogram. This resolves promptly with discontinuation of the drug. Another atypical side-effect is described as weakness or asthenia. No metabolic, hepatic or blood-related adverse events have been identified. Therefore, monitoring of complete blood counts, electrolytes and liver function tests is not clinically warranted. Tiagabine does not impact on the metabolism of other drugs. Other anti-epileptic drugs can impact on the metabolism of tiagabine.

Levetiracetam

Levetiracetam is a broad-spectrum drug that has undergone extensive testing. Therefore, more syndromes have been explored than for some of the other newer anti-epileptic drugs. It is effective in partial seizures. A study of patients with newly diagnosed partial or generalized tonic-clonic seizures showed no differences in efficacy or tolerability between levetiracetam and carbamazepine. Levetiracetam is the only newer anti-epileptic drug approved for use in juvenile myoclonic epilepsy. It was also effective for idiopathic generalized tonic-clonic seizures in an add-on situation. Case series have indicated efficacy in other idiopathic epilepsy syndromes.

One reason levetiracetam has become a very popular choice, for both initial and add-on therapy, is that it is easy to use. It can be started at a therapeutic dose, and data indicate that onset of action is within a day. Patients can be started on either 500mg once daily or 500 mg twice daily. The dose can then be increased gradually as necessary, with maximal dose typically being 3000 mg/day. If necessary, it can be started at a higher dose, although it is not clear if this is clinically necessary. In one study, levetiracetam was started at 4000 mg/ day, and was well tolerated. Serum levels are typically within the range of 10—40mgA, but will vary substantially over the course of a day, because of the short half-life of the drug. Another advantage of levetiracetam is that it is associated with no drug interactions because it is predominantly renally excreted, and shows limited metabolism in humans. As with other renally excreted drugs, lower doses should be used in the elderly, because they have a reduced creatinine clearance .

Side-effects of levetiracetam include irritability, somnolence, dizziness, asthenia and headache. Other uncommon adverse events include behavioural problems, depression and psychosis. Levetiracetam does not often cause rash, and is therefore a reasonable choice in patients with a history of hypersensitivity syndrome. To date, there has been no indication that levetiracetam can cause idiosyncratic safety problems such as aplastic anaemia or hepatic failure. Therefore, routine monitoring of liver function tests, blood counts and electrolytes is not clinically warranted.

Levetiracetam was recently approved in an i.v. formulation. It is being used more frequently for inpatients with other medical conditions who require immediate anti-epileptic drug therapy, and in status epilepticus, but no formal studies have been done.

Zonisamide

Zonisamide is felt to be broad spectrum. Unfortunately, trials in generalized seizure syndromes have not been performed to confirm this clinical impression. The only randomized controlled trials were performed in adults with partial seizures. Randomized trials in patients with newly diagnosed epilepsy have also not been done. Case series and open studies support a role for zonisamide in the treatment of syndromes associated with myoclonus, including juvenile myoclonic epilepsy and progressive myoclonic epilepsies. Zonisamide is better tolerated when it is titrated at initiation. It can be started at 50mg/day or 100 mg/day. The dose should then be titrated by 50mg/week or 100 mg every other week. Doses of 200-300 mg are common, and up to 500 mg can be necessary in difficult-to-treat patients. Serum concentrations of 20-40 mg/1 are considered therapeutic. Since the half-life is long (up to 60 h), serum concentrations can be expected to be steady over the course of the day. Common dose-related adverse events include fatigue, weight loss, dizziness, somnolence, anorexia and abnormal thinking, and decreased sweating. Zonisamide has also been associated with idiosyncratic side-effects. These include hypersensitivity syndromes such as rash and Stevens-Johnson syndrome and renal calculi. Patients living in hot climates, who are on high doses and have a family history of renal calculi may be at higher risk. Patients receiving zonisamide should be advised to drink plenty of fluids. The recurrence rate, even if zonisamide is continued, is not 100%. Hypohidrosis has also been seen, again more common in hot climates, and may infrequently lead to heat stroke in children. Zonisamide is well absorbed and is not extensively bound to plasma proteins. Since enzyme-inducing drugs such as carbamazepine and phenytoin can significantly reduce the half-life of zonisamide, higher doses will be needed in patients taking them in combination. Zonisamide metabolism can also be inhibited by a number of medications as well as by grapefruit juice. Zonisamide has essentially no impact on the pharmacokinetic parameters of other drugs.

Pregabalin

Pregabalin has a similar efficacy profile to gabapentin, that is to say its use is restricted to partial seizures. No formal monotherapy studies have been completed, in either newly diagnosed or refractory patients. An advantage of pregabalin when compared to gabapentin, is that it is highly bioavailable, and does not require active dose-dependent transport in the Gastrointestinal tract. Pregabalin is better tolerated when titrated. A recent study indicated that starting at the highest dose of 600 mg, while safe, led to 32% dropouts. Pregabalin can be initiated at 50-75 mg twice daily, and titrated by 50-75 mg every week or two weeks. There are very few clinical data available regarding the safety and tolerability of doses above 600mg/day. Twice-daily dosing is commonly used, despite the short half-life of the drug. A study comparing efficacy and tolerability of the same total daily dose, given as twice daily or thrice daily showed no statistically significant difference, but there was a trend to both better efficacy and tolerability when thrice daily was used. Therefore, it is reasonable to start off with a twice-daily regimen, and switch to thrice daily only if side-effects are present and/or breakthrough seizures are occurring. Pregabalin levels have only recently become available, and their clinical utility is unknown. Common dose-related side-effects predominantly affecting the central nervous system include dizziness, somnolence and ataxia. Weight gain and peripheral oedema are also seen. Weight gain is also dose related. To date, no idiosyncratic side-effects have been identified. Need for routine monitoring of liver functions, electrolytes and blood counts is therefore not established. Pregabalin is almost exclusively excreted unchanged in the urine, and does not undergo metabolic changes. To date, no drug-drug interactions have been uncovered, and none are expected.

Vigabatrin (not approved in the usa)

Vigabatrin, like felbamate, is an anti-epileptic drug that has been associated with a significant adverse drug effect (irreversible visual field restriction) that limits its use to those patients who have severe epilepsy which has not responded to other anti-epileptic drugs. In adults, vigabatrin appears to have a narrow spectrum of action, and its use is primarily restricted to those patients with refractory partial epilepsy. It has been found inferior in efficacy to carbamazepine in trials of newly diagnosed patients. It has also been known to worsen myoclonus. In infants, however, vigabatrin was found to be highly effective in the devastating childhood epilepsy known as infantile spasms, or West syndrome [99]. In a randomized trial, there was a 78% reduction in seizures on vigabatrin, compared with 26% on placebo. Moreover, in the open-label phase, 38% of children were completely spasm free. Vigabatrin is particularly effective in patients with spasms associated with tuberous sclerosis. Vigabatrin can be initiated at 500-1000mg, given once or twice a day. The dose can be increased in 500-1000 mg weekly increments, up to 3000 mg. Some patients worsen at higher doses. Vigabatrin has unique pharmacokinetic properties, in that its mechanism of action involves changes in brain chemistry that far outlast its presence in the bloodstream. Therefore, vigabatrin levels are not useful for therapeutic monitoring.

As noted, the main safety concern relates to irreversible peripheral visual field defects. This problem is estimated to occur in 30-50% of patients receiving the drug. Often, this will be picked up on screening visual field testing, but the patients will not spontaneously report any problem. Since early development of this problem has not been seen in controlled studies, there may be a 3-month ‘window of opportunity’ to see if the drug works before there is a risk of the visual field defect. However, once it occurs, it is not reversible.

Other side-effects of vigabatrin include drowsiness, depression, weight gain, dizziness and rare psychosis .

Physicians may be confused about whether measurement of serum concentrations is important for epilepsy patients, particularly for the new anti-epileptic drugs. Most experts believe that it is more important to individualize dosing, and to understand the serum concentration that is optimal on an individual basis. This may mean that some individuals have optimal control and side-effect profile at a serum concentration that is outside of the recognized therapeutic norms.

Approximately 25% of patients will achieve seizure freedom below the recognized therapeutic range, and about the same number will require serum concentrations above the range to obtain the maximum seizure benefit, often with few side-effects.

Many of the new anti-epileptic drugs are considered to have a wide therapeutic range. The range of serum concentrations that can be beneficial varies widely from one individual to another, yet for a particular individual it may be very important to maintain serum concentrations with minimal variability. For example, one patient on lamotrigine may have excellent seizure control within a range of 2-4 mg/1, but may experience dizziness and diplopia if the level increases to 5 mg/1. Another patient, in contrast, may have breakthrough seizures if their level drops below 7mg/l, but may tolerate levels up to 10mg/l.

Another area of confusion relates to timing of levels. Some anti-epileptic drugs have short half-lives, but nonetheless are dosed two times a day. In some cases, there is reasonably good scientific evidence that the drug has a longer duration of action in the central nervous system than would be expected based on its half-life. This has led to the common practice of dosing these drugs two times a day. It seems that other anti-epileptic drugs must be present at relatively constant serum concentrations throughout the day, to prevent seizure breakthroughs. Carbamazepine and phenytoin seem to fall in this category. What does this mean in terms of levels? For drugs that are dosed less frequently than their half-lives, the levels will be expected to vary substantially over the course of a day. Thus, levels can only be compared if they are measured at the same time of day. Take, for example, a patient who is receiving levetiracetam 1000 mg two times a day, at 8 a.m. and 8 p.m. He always gets his serum levels measured at his physician’s office at the time of his appointments, 2h after his dose. The levels would therefore be close to the peak. When measured, they are 20-25 mg/1. The patient now has a seizure at 7 p.m. and the level in the emergency room is 10 mg/1. One interpretation would be that the patient is non-compliant or the levels have dropped, but this would be incorrect. In fact, the levels were measured at trough, and are exactly where they should be. Now consider a patient on sustained-release carbamazepine, who receives 400 mg two times a day, at 8 a.m. and 8 p.m.. In this case, serum concentrations would be expected to vary very little over the course of a day, and a treating physician will have to investigate the cause of any major fluctuations. This can be very confusing for a treating physician. Nonetheless, levels can be useful if employed properly.

The following strategy may be recommended for using serum levels:

•   When initiating therapy, increase doses slowly, until seizure control is achieved or side-effects occur. If seizures are controlled, obtain a serum concentration at that time. This will identify the patient’s ‘therapeutic optimal level’.

•   Measure all levels at the same time of day, if possible. This will provide some indication of compliance, and also of variability in a given patient.

•   For drugs with sustained-release formulations, or long half-lives, serum concentrations should remain within a small range (no more than 20-25% change), even if obtained at different times of day, once a steady state is achieved (e.g. phenobarbital, phenytoin (administered two times a day), Tegretol XR™ or Carbatrol™ (sustained-release carbamazepine formulations), zonisamide, Depakote ER™ (sustained-release valproate formulation), topiramate and lamotrigine (administered two times a day in patients who are not receiving enzyme-inducing anti-epileptic drugs). If a breakthrough seizure occurs in the setting of a low serum concentration, consider non-compliance or inappropriate dosing schedule.

•   For drugs with short half-lives and no sustained-release formulation, serum concentrations should be consistent when taken at trough, but may vary by >50% if taken at other times, and should not be interpreted as indication of non-compliance (e.g. non-sustained-release valproate, gabapentin, levetiracetam, carbamazepine [administered two times a day], tiagabine and pregabalin).