The pharmacological treatment of epilepsy in adults

The pharmacological treatment of epilepsy in adults
Lancet Neurol 2011; 10: 446–56
Treatment decisions in epilepsy need to be individualised on the basis of careful analysis of the risk-benefi t ratio of
Clinical Pharmacology Unit,
each available option. Key decision steps include the time at which antiepileptic drug treatment should be started,
Department of Internal
which drug should be chosen for fi rst-line therapy, and which strategy is most appropriate for people who did not
Medicine and Therapeutics,
respond to the initially prescribed drug. With more than 20 antiepileptic drugs currently available to treat epilepsy in
University of Pavia, Pavia, Italy
adults, opportunities to tailor drug therapy have never been greater, but optimum use of such a complex
(Prof E Perucca MD); Institute of
Neurology, Istituto di
armamentarium is a challenge even for the epilepsy specialist. Antiepileptic drug choice is primarily based on
Ricovero e Cura a Carattere
evidence of effi
cacy and eff ectiveness for the individual’s seizure type, but other patient-specifi c factors need to be
Scientifi co (IRCCS), C Mondino
considered, including age, sex, childbearing potential, comorbidities, and concomitant medications.
Foundation, Pavia, Italy
(E Perucca); and Department of
Clinical Neuroscience,
Introduction
Karolinska Institutet,
10% of people will have at least one seizure in their against potential adverse eff ects, while at the same time Stockholm, Sweden
lifetime, and about a third of them will go on to develop considering the risks associated with withholding of epilepsy.1,2 In terms of both prevalence and cumulative treatment. These risks should be assessed in the broadest incidence, epilepsy is one of the most common serious sense, taking into account the patient’s perspective in neurological disorders, with the same burden of disease relation not only to the risk of further seizures but also as lung cancer in men or breast cancer in women.3 It has the risk of seizure-related morbidity and mortality, as been estimated that, in Europe alone, the costs attributable well as AED toxicity. Ideally, assessment of the potential emilio.perucca@unipv.it
to active epilepsy exceed €20 billion per year.4 In addition benefi ts of treatment should involve an understanding of to psychosocial disability and seizure-related injuries, the natural course of untreated epilepsy and its epilepsy is associated with other comorbidities, including consequences. However, understanding of the natural depression and increased mortality.5 In a population- cult to achieve because treatments have based cohort6 that was followed up for 40 years since been available since the 19th century and it is considered childhood, overall mortality in people with epilepsy was unethical to leave active epilepsy untreated. SUDEP is 24%—three-times the rate expected in the general the most devastating threat, although available data population—and more than half the deaths were related suggest that such deaths rarely occur with the fi rst few to epilepsy itself, including sudden unexpected death seizures.11 Seizures can also impose restrictions on (SUDEP) in a third of all fatalities.
everyday activities such as driving, and can restrict career Because of the seriousness of the disorder and its choices, all of which have far greater relevance in adults epidemiological dimension, it is of concern that epilepsy is often suboptimally diagnosed and managed, even in The risk of recurrence after a fi rst unprovoked seizure the European region.7 Conceptual and practical has been analysed in several studies. A meta-analysis12 considerations for the medical management of epilepsies from 1991 concluded that the average risk of having a in children were reviewed in this journal in 2008.8 further seizure was 46%, but variation across studies was However, most people with epilepsy are adults, with substantial. The risk of subsequent seizures was much features that are specifi c to their age group. Studies in higher after a second seizure (more than 70%), even in a recent years have provided new grounds for rational cohort in which most patients were on treatment after management of these patients. Here, we focus on the their second seizure.13 On the basis of this increased risk, pharmacological treatment of this disorder in adults by there has been a consensus that treatment is indicated addressing a number of key questions, including after at least two unprovoked seizures.
indications for starting treatment, initial drug selection, From a practical viewpoint, a key fi nding is that the risk strategies when treatment fails, and risks and benefi ts of of recurrence after a fi rst seizure varies considerably in withdrawal of treatment in seizure-free patients.
relation to the patient’s characteristics. In individuals with risk factors (eg, epileptiform electroencephalogram When should treatment be started?
[EEG] discharges, an abnormal neurological examination, Antiepileptic drugs (AEDs) are the mainstay of the or other evidence of a structural CNS abnormality treatment of epilepsy, and although their number has presumed to be responsible for the seizure), the risk of expanded exponentially, current principles governing recurrence after a fi rst seizure might be similar to the drug therapy are in many ways similar to those average risk after two seizures.12 This has been taken into established a century ago.9 Because AED therapy is account in the newly proposed International League typically maintained for several years and often for life, Against Epilepsy (ILAE) defi nition of epilepsy as “a particularly in adults, a decision to initiate treatment has disorder of the brain characterized by an enduring far-reaching consequences and needs to be based on predisposition to generate epileptic seizures”14 and careful risk-benefi t analyses.10 requiring “occurrence of at least one unprovoked seizure”, www.thelancet.com/neurology Vol 10 May 2011
by contrast with the old defi nition requiring at least two recurrence and, accordingly, the size of the treatment seizures.15 Although the decision to treat is distinct from eff ect.21 Patients who presented with one seizure only the issue of the defi nition of epilepsy, the newly proposed were classed as low risk, those with two or three seizures, defi nition is congruent with the notion that there could a neurological disorder, or an abnormal EEG were classed be situations in which the recurrence rate after one as medium risk, and those with two of these features or seizure is so high that treatment can be justifi ed.
more than three seizures were classed as high risk. Evidence that treatment can be eff ective in prevention Recurrence rates at 2 years ranged from 30% in the low- of recurrence after a single seizure has been provided by risk group to 73% in the high-risk group. Immediate two large open-label, pragmatic randomised controlled treatment was of little or no benefi t in the low-risk group, trials. In the First Seizure Trial Group (FIRST) study16 but had potentially worthwhile benefi ts in the other and the Multicentre Study of Early Epilepsy and Single groups. In the high-risk group, in particular, the Seizures (MESS) study,17 patients were randomly allocated probability of recurrence at 2 years was reduced from to immediate treatment or to deferred treatment. 73% to 50%. These data can be helpful in deciding 419 adults and children were enrolled in the FIRST trial whether to treat or not to treat.
within a week of a fi rst unprovoked tonic-clonic seizure. The severity of the seizures, their timing (during sleep Of the patients randomised to immediate treatment, 24% or while awake), and individual vulnerabilities might also had a recurrence within the fi rst 2 years compared with contribute to the decision of whether treatment is 42% of patients who were untreated until recurrence.16 indicated. The following recommendations are largely in The MESS study enrolled adults and children in whom agreement with those made by Marson.22 First, treatment there was uncertainty about the indication to start is indicated after two seizures if their severity is of treatment, including 812 individuals who had had one signifi cance for the individual (for some patients, by seizure, mostly tonic-clonic in nature. At 2 years after contrast, occasional focal seizures with subtle sensory randomisation, 32% of those on treatment and 39% of symptoms might not have an eff ect on quality of life that those untreated had had seizure recurrence.17 Most justifi es treatment), if the interval between them is not importantly, both studies indicated that deferral of more than 1 or 2 years, and if the informed patient so treatment had no eff ect on long-term prognosis. In the wishes. Second, because about 50% of individuals with a FIRST trial, the 2-year remission rate for patients treated fi rst unprovoked seizure who are not treated will never after a fi rst seizure (60%) was equivalent to that for have a second seizure, and because AED treatment patients whose treatment was initiated only after carries a substantial risk of adverse eff ects, it is reasonable recurrence (59%).18,19 In the MESS trial, 92% of patients to defer treatment until after seizure recurrence in most in the immediate treatment group and 90% of those in cases. However, early treatment might be justifi ed in the deferred treatment group achieved 2-year remission patients with a high recurrence risk, in particular if the after 5 years of follow-up. Moreover, in the MESS study consequences of further seizures are expected to be there were no diff erences in quality-of-life outcomes severe. Patients with a fi rst unprovoked seizure after a between the two groups at 2 years except for the eff ect on stroke or a seizure caused by other identifi able lesions ability to drive, in which patients randomly allocated to have a high recurrence rate, and fragile elderly patients deferred treatment were disadvantaged.20 Similarly, there are also likely to be more vulnerable if seizures recur.
was no evidence of death, injuries, or status epilepticus As always, the doctor’s task is to describe the diff erent occurring more frequently in patients allocated to management options and their likely consequences.
deferred treatment.17 Interestingly, when asked, more The fi nal decision to accept treatment belongs to the
patients in the immediate treatment group (22%) informed patient.
expressed preference for the other treatment policy than
did those allocated to deferred treatment (5%), possibly Which AED should be chosen for initial
refl ecting concerns about adverse drug eff ects. Notably, treatment?
however, patients were recruited to the MESS study only The ultimate goal of epilepsy treatment is lasting freedom
if there was uncertainty about the indication to start from seizures without adverse eff ects. Therefore,
treatment, and therefore the results from this study selection of the fi rst AED should be guided primarily by
might not necessarily be applicable to the majority of the
cacy for the patient’s seizure type or population with newly diagnosed epilepsy.
epilepsy syndrome and by tolerability considerations, Although immediate treatment after a fi rst seizure preferably on the basis of data from well designed reduced the recurrence rate in both studies,16,17 the randomised controlled trials. The proportion of patients magnitude of risk reduction at 2 years diff ered between who remain on the allocated AED for a period of time, the FIRST (42% reduction) and MESS (18% reduction) often referred to as eff ectiveness, provides a combined trials. The discrepancy is likely to be related at least in measure of effi part to diff erences in risk factors between the two parameter should not be overestimated because not all populations. Indeed, a reanalysis of the MESS data adverse eff ects that aff ect quality of life necessarily lead identifi ed several factors that predicted the risk of to discontinuation of the off ending drug. A systematic www.thelancet.com/neurology Vol 10 May 2011
Academy of Neurology (AAN) guidelines25 list many old SIGN‡ ILAE§
and new AEDs without expressing any preferences among them; the UK National Institute for Health and Clinical Excellence (NICE)26 recommends preferential use of older agents unless there are specifi c reasons for doing otherwise; the Scottish Intercollegiate Guideline Network (SIGN)27 identifi es specifi cally carbamazepine, valproate, lamotrigine, and oxcarbazepine as fi rst-line agents; and the ILAE rates phenytoin and carbamazepine as the AEDs with the highest quality of evidence for cacy and eff ectiveness.23 Phenytoin, however, is rarely used in Europe at present, mainly owing to its complicated AED=antiepileptic drug. AAN=American Academy of Neurology. NICE=UK National Institute for Health and Clinical pharmacokinetics and spectrum of adverse eff ects. This Excellence. SIGN=Scottish Intercollegiate Guideline Network. ILAE=International League Against Epilepsy. *Patients “can be initiated on standard AEDs such as carbamazepine, phenytoin, valproic acid, phenobarbital or on the new AEDs adverse-eff ect and pharmacokinetic profi le explains why lamotrigine, gabapentin, oxcarbazepine, or topiramate. Choice of AED will depend on individual patient phenytoin was excluded from the list of fi rst-line drugs characteristics.”25 †“The newer AEDs gabapentin, lamotrigine, levetiracetam, oxcarbazepine, tiagabine, topiramate and vigabatrin, within their licensed indications, are recommended for the management of epilepsy in people who have not benefi ted from treatment with the older antiepileptic drugs such as carbamazepine or sodium valproate, or for Two large randomised controlled trials in newly whom the older AEDs are unsuitable because there are contraindications to the drugs; they could interact with other diagnosed epilepsy have been published since 2006. A drugs the person is taking (notably oral contraceptives); they are already known to be poorly tolerated by the double-blind trial meeting ILAE class I criteria showed individual; the person is a woman of childbearing potential.”26 ‡Carbamazepine, sodium valproate, lamotrigine and oxcarbazepine can all be regarded as fi rst-line treatments for partial and secondary generalised seizures. Sodium valproate and lamotrigine are drugs of choice for primary generalised seizures and should also be prescribed if there is levetiracetam and controlled release carbamazepine in any doubt about the seizure types and/or syndrome classifi cation. The side eff ect and interaction profi les should direct adults with focal seizures.28 The pragmatic Standard and the choice of drug for the individual patient.”27 §The International League against Epilepsy guideline23 rated AEDs only New Antiepileptic Drugs (SANAD) trial,29 which enrolled on the basis of the quality of evidence for effi cacy or eff ectiveness (level A highest). Vigabatrin also received a C rating although it is not recommended as initial treatment in patients with focal seizures owing to its ocular toxicity. a mixed population of adults and children, reported ¶A subsequently completed double-blind trial showed that levetiracetam meets ILAE class I criteria for non-inferiority lamotrigine to be the most eff ective compared with carbamazepine, oxcarbazepine, topiramate, and Table 1: AEDs recommended by major current guidelines as possible fi rst-line therapy for adults with
gabapentin in patients with focal seizures, and valproate focal seizures, with or without secondary generalisation
to be more eff ective than lamotrigine and topiramate in patients with generalised and unclassifi able epilepsies.30 In this group, the diff erence in favour of valproate was review of controlled trials comparing effi cacy and even larger in the subgroup of patients with idiopathic eff ectiveness of AEDs in newly diagnosed epilepsy generalised epilepsy. Although the evidence from SANAD concluded that there are major weaknesses in the quality might be questioned because of its open-label design and of the available evidence.23 Indeed, of 33 eligible trials in other potential biases,31 its multigroup comparisons with adults with focal seizures, only two were rated as class I fl exible dosage and a follow-up period of up to 6 years (the highest rating in terms of quality of evidence), one makes it a unique and valuable study.
was rated class II, and 30 received the lowest rating There are several properties in addition to effi (class III). None of the trials in adults with generalised eff ectiveness that might aff ect the choice of AEDs and that tonic-clonic or other generalised seizure types achieved typical randomised controlled trials might not capture. class I or II ratings. Evaluation of the evidence is further These include adverse eff ects such as rare idiosyncratic complicated by the fact that the design of many trials reactions, teratogenic eff ects, and chronic side-eff ects. comparing older and newer drugs incorporated bias in Enzyme-inducing eff ects and potential for drug interactions the choice of formulation, target dosages, or titration are also important, as are the availability of parenteral rates, which favoured the sponsor’s product.24 Additionally, formulations and the possibility of rapidly reaching an most randomised controlled trials are regulatory trials effi cacious target dose in some cases. Indeed, all existing designed to obtain a marketing licence. Although such guidelines emphasise the need to consider individual studies might provide evidence of effi patient characteristics when selecting an AED. In addition on highly selected patients often treated with fi xed doses to seizure types (panel), characteristics of special relevance for a comparatively short duration. This setting is far include childbearing potential, old age, and comorbidities.
from routine clinical practice, which involves fl exible Several observational studies have reported higher rates long-term treatment of heterogeneous populations.
of major congenital malformations with use of valproate In view of these shortcomings, it is perhaps not during pregnancy compared with other AEDs such as surprising that present major guidelines,23–27 all of which carbamazepine or lamotrigine.33–39 A recent population-were published between 2003 and 2006, diff er in their based case-control study confi rmed that, compared with recommendations of specifi c AEDs for fi rst-line other AEDs, valproate carries an increased risk of spina monotherapy (table 1). For focal seizures—by far the bifi da, atrial septal defects, cleft palate, and most prevalent seizure type in adults—the American craniosynostosis.40 Additionally, observational studies41 www.thelancet.com/neurology Vol 10 May 2011
indicate that fetal exposure to valproate can adversely aff ect cognitive development. Children of mothers who Panel: Effi
cacy spectrum of the main antiepileptic drugs in diff erent seizure types in
were treated with valproate during pregnancy had signifi cantly lower IQ (although within the normal range) Eff ective against focal seizures and most generalised seizure types
at the age of 3 years compared with children exposed to carbamazepine, lamotrigine, or phenytoin.41 The • Benzodiazepines*teratogenic eff ects of valproate seem to be dose-dependent and more prominent at doses of 800 mg per day or more.34–36,41–44 Valproate is thus not the drug of choice for women of childbearing potential, particularly at doses above 800 mg per day. For focal seizures, for which many treatment options are available, valproate should probably be reserved as a third or lower choice in women of childbearing potential. For primarily generalised seizures, the number of AEDs that could be used as alternatives to valproate is limited (panel). Lamotrigine is a reasonable Primarily eff ective against focal seizures, with or without secondary generalisation
especially in syndromes associated with myoclonic manifestations or absences.45,46 Treatment with lamotrigine during pregnancy is also complicated by pronounced changes in its plasma concentrations, which might require frequent dose adjustments.47 Levetiracetam and topiramate are other alternatives for women of childbearing potential with generalised epilepsy. Clinical data for the teratogenic risks with levetiracetam are still insuffi levetiracetam during pregnancy can also be complicated Eff ective against absence seizures
by a decline in its plasma concentrations.48 Plasma topiramate also decreases during pregnancy,49 and we do not consider topiramate a fi rst-choice alternative because Modifi ed from Shorvon and colleagues,32 by permission of Wiley-Blackwell. *Occasionally exacerbate tonic seizures, particularly after intravenous use in patients with Lennox-Gastaut syndrome. †Not eff ective against absence seizures. of its adverse eff ects on cognition50 and because preliminary ‡Can aggravate myoclonic seizures in some patients; the effi cacy of lamotrigine is best documented against focal and data51 for its potential teratogenicity raise some concerns. secondarily generalised tonic-clonic seizures, primarily generalised tonic-clonic seizures, absence seizures, and drop attacks Uncontrolled observational and retrospective studies52 associated with Lennox-Gastaut syndrome. §Effi cacy against tonic and atonic seizures has not been documented; the effi of levetiracetam is best documented against focal and secondarily generalised tonic-clonic seizures, primarily generalised have suggested that zonisamide can also be eff ective in tonic-clonic seizures, and myoclonic seizures. ¶Effi cacy against absence seizures has not been documented; the effi primarily generalised seizures, but there are no data on topiramate is best documented against focal and secondarily generalised tonic-clonic seizures, primarily generalised the safety and pharmacokinetics of this drug during tonic-clonic seizures, and drop attacks associated with Lennox-Gastaut syndrome. ||Effi cacy of zonisamide is best documented against focal and secondarily generalised cacy against absence and primarily generalised tonic-clonic seizures has not been documented; The incidence of epilepsy increases exponentially in cacy of rufi namide is best documented against focal and secondarily generalised tonic-clonic seizures, and drop attacks associated with Lennox-Gastaut syndrome. ††Effi cacy against absence, myoclonic, and primarily generalised old age, and elderly people are more susceptible to the tonic-clonic seizures has not been documented; the effi cacy of felbamate is best documented against focal and secondarily adverse eff ects of AEDs.53 Two randomised controlled generalised tonic-clonic seizures, and drop attacks associated with Lennox-Gastaut syndrome. ‡‡Carbamazepine, phenytoin, trials in epilepsy with onset in old age reported that cacious against primarily generalised tonic-clonic seizures. Carbamazepine, phenytoin, oxcarbazepine, tiagabine, and, presumably, eslicarbazepine acetate can precipitate or aggravate absence and myoclonic carbamazepine was associated with high rates of seizures. §§Gabapentin and pregabalin can precipitate or aggravate myoclonic seizures. ¶¶Tentative classifi cation; discontinuation due to adverse eff ects compared with lacosamide has not been assessed in patients with primarily generalised seizures. ||||Can precipitate or aggravate myoclonic cacious against myoclonic seizures.
54,55 and gabapentin.55 The poor outcomes with carbamazepine could be explained in part by use of immediate-release formulations rather than sustained- impaired bone health, endocrine dysfunction, and, release formulations54,55 and, in one trial, by selection of possibly, changes in cholesterol concentrations and other an excessively high target dose of 600 mg per day.55 In a markers of vascular risks.58 The use of enzyme inducers smaller randomised controlled trial in elderly patients, such as carbamazepine, phenytoin, or phenobarbital as outcome with sustained-release carbamazepine at 400 mg fi rst-line drugs has therefore been questioned, particularly per day was as good as with lamotrigine 100 mg per day.56 after the introduction of alternative medications that are Irrespective of these fi ndings, however, carbamazepine is devoid of enzyme-inducing activity (eg, levetiracetam) or not an ideal fi rst-line AED in elderly people because its have a reduced interaction potential (eg, lamotrigine and enzyme-inducing properties can cause adverse oxcarbazepine).58 Because of the low cost and other interactions with comedications that are frequently taken carbamazepine, this drug should still be regarded as a In addition to causing adverse drug interactions, enzyme reasonable fi rst choice for patients with focal seizures; induction contributes to chronic adverse eff ects such as however, other drugs, particularly lamotrigine or www.thelancet.com/neurology Vol 10 May 2011
Advantages
Disadvantages
Carbamazepine High-level evidence of effi cacy, extensive experience, low cost, mood Enzyme inducer, high interaction potential, hyponatraemia Non-enzyme inducer, no interactions, rapid titration, eff ective in cacy (except in the elderly), weight gain cacy in the elderly, non-enzyme inducer, Slow titration; dose requirements infl uenced by interactions with valproate, enzyme inducers, and oestrogen-containing contraceptives cacy, rapid titration, non-enzyme inducer, Low potential for enzyme induction, lower risk of rashes than for Higher risk of hyponatremia than for carbamazepine, reduces blood concentrations of oral contraceptive steroids Extensive experience, very low cost, once-daily dosing Enzyme inducer, high interaction potential, cognitive and behavioural adverse eff ects cacy, rapid titration, extensive experience, Complicated pharmacokinetics, enzyme inducer, high interaction Eff ective for migraine prophylaxis, low potential for enzyme induction Slow titration, cognitive adverse eff ects Enzyme inhibitor, some interaction potential, weight gain, teratogenic potential Table 2: Selected major advantages and disadvantages of potential fi rst-line antiepileptic drugs for treatment of adult-onset focal seizures
bipolar depression,62 and pregabalin is eff ective in relieving Advantages
Disadvantages
the symptoms of generalised anxiety disorder.63 Slow titration; dose requirements infl uenced by The presentation of epilepsy can sometimes require interactions with valproate, enzyme inducers, rapid institution of a fully eff ective therapy. If that is the and oestrogen-containing contraceptives; can aggravate myoclonic seizures in some patients case, lamotrigine or topiramate, which require slow titration over several weeks, are undesirable choices and AEDs such as levetiracetam, valproate, or phenytoin, Eff ective for migraine prophylaxis, low Slow titration, cognitive adverse eff ects which can be started at (or titrated rapidly to) a fully eff ective dose, are preferred.64 These drugs can also be Enzyme inhibitor, some interaction potential, given as intravenous loading doses if needed.
Finally, the cost of medications and reimbursement Table 3: Selected major advantages and disadvantages of potential fi rst-line antiepileptic drugs for
considerations also aff ect the choice of AED. Most people treatment of adult-onset primarily generalised seizures
with epilepsy live in developing countries where phenobarbital might be the only aff ordable medication and there are strong arguments for promotion of its use.65 levetiracetam, are preferable when drug interactions or Tables 2 and 3 show advantages and disadvantages of enzyme induction are foreseen as signifi cant problems, the most commonly used AEDs for fi rst-line treatment of especially in patients on other medications or those for focal and generalised seizures.
whom bone health is a concern.
Comorbidities need to be taken into consideration in the Which formulation?
choice of AEDs not only because of the interactions with Some AEDs are available as sustained-release medications used to treat them, but also because some formulations. At least for carbamazepine, sustained-AEDs might adversely or positively aff ect comorbid release tablets off er superior tolerability to immediate-disorders. Valproate and topiramate, for example, are release formulations when a twice daily schedule is used.66 eff ective for migraine prophylaxis59 and might be The same evidence is not available for most other AEDs, considered when epilepsy and migraine coexist. Psychiatric and the introduction of modifi ed-release formulations is comorbidities, especially depression, are common in often motivated by marketing considerations rather than people with epilepsy60 and there has been controversy demonstration of any clinical advantage.
about the possibility that AEDs might increase the risk of For many AEDs, there is also a choice between trade suicidal ideation, suicide attempt, and completed suicide. names and generics. A heated debate about the safety of Although this issue remains unresolved,61 it is reasonable generic substitution in epilepsy is ongoing.67 A recent to be cautious with levetiracetam, which is associated with systematic review of randomised controlled trials an augmented incidence of psychiatric adverse events, as comparing trade name and generic AEDs noted no a fi rst-choice AED in people with psychiatric comorbidities. evidence for an association between loss of seizure Conversely, carbamazepine and valproate can be used as control and generic substitution,68 although these results mood stabilisers, lamotrigine has benefi cial eff ects in should be interpreted cautiously owing to the small www.thelancet.com/neurology Vol 10 May 2011
sample size and short duration of assessment. Although concentrations in the individualisation of therapy is generic substitution does not apply at the time epilepsy discussed extensively in a recent ILAE guideline.70treatment is started, the likelihood of continuous supply of a product from the same manufacturer when selecting What is the next course of action when initial
a generic drug needs to be considered.
monotherapy fails?
About 50% of a typical adult population will achieve
Which dosage?
sustained seizure freedom without intolerable side-eff ects Correct dosage is as important as choice of the most on the initially prescribed AED.71 If the initial monotherapy appropriate drug. For most AEDs, a gradual dose titration fails, subsequent management should take into account a can improve CNS tolerability, reduce the risk of number of factors. If the fi rst AED had to be discontinued idiosyncratic adverse reactions, or both.64 Therefore, because of an idiosyncratic reaction, an alternative AED unless an immediate anti-seizure eff ect is required, should be tried. Care should be taken to avoid, if possible, treatment is generally started with a low dose, which is drugs that are likely to show crossreactivity for the same then increased with time. The optimal duration of the reaction. For example, if an individual had a rash in titration period varies with type of AED, selected target response to carbamazepine, the probability of the rash maintenance dose, and individual response (table 4).
recurring with another AED has been reported to be 58% Ideally, to minimise toxic eff ects, AEDs should be given for phenytoin, 33% for oxcarbazepine, 27% for at the lowest dose that provides sustained freedom from phenobarbital, and 20% for lamotrigine.72seizures. Because such doses vary between individuals, it If failure of the fi rst AED was caused by a lack of is usually advisable to aim initially at the lowest effi cacy, with seizures persisting after titration to the maintenance dose that is expected to be eff ective in the highest tolerated dose, the fi rst step is to exclude non-majority of patients. In recent years, evidence has compliance and to reassess the diagnosis and the accumulated that people with newly diagnosed epilepsy appropriateness of the initial treatment. In a report from respond to relatively low doses of the initially prescribed the UK, 42 (19%) of 222 individuals on AED treatment AED. Specifi cally, most patients achieving seizure diagnosed as having epilepsy by a non-specialist were freedom on the fi rst AED do so at doses of 400 mg per day reported to have their diagnosis questioned after review for carbamazepine,28 1000 mg per day for levetiracetam,28 by a specialist service.73 Non-compliance is a major reason 125–200 mg per day for lamotrigine,69 and 600–1000 mg for treatment failure and can have serious consequences per day for valproate.69 The initial target maintenance beyond seizure control: mortality has been shown to be dosage should also take into account the characteristics of signifi cantly increased in patients with epilepsy during the individual (eg, elderly patients might need reduced periods of non-adherence to prescribed treatment.74 If a dosages) and their attitudes towards potential side-eff ects change in AED treatment is indicated, the conventional and risk of seizure recurrence. Should seizures recur at recommendation is to switch gradually to monotherapy the selected initial maintenance dose, the dose can be with another drug.26 This approach is supported by increased according to clinical response.
evidence that up to 20–30% of individuals who are For old and new AEDs, reference ranges of serum resistant to the initial AED achieve seizure freedom on concentrations that are often associated with an optimal an alternative monotherapy,71,75,76 and by results of a small response have been proposed.70 Aiming at an initial pragmatic randomised controlled trial in which target serum concentration rather than initial target alternative monotherapy was not inferior to add-on maintenance dose might be a reasonable option, therapy in patients previously unexposed to AED particularly with phenytoin, which shows highly combinations.77 Other authors, however, feel that variable and dose-dependent pharmacokinetics. combination therapy could be tried earlier, particularly in However, there is a high variability in the serum severe epilepsies when the fi rst AED seems to have been concentrations associated with the best response, and partially eff ective and well tolerated, and the probability newly diagnosed patients often achieve seizure freedom of seizure freedom with monotherapy is regarded as at concentrations below the lower limit of the reference low.78 This approach might be further justifi ed by evidence range. Modern use of drug monitoring underemphasises that there seem to be specifi c AED combinations that can the value of reference ranges and relies increasingly on be associated with especially pronounced effi the identifi cation of the serum concentration that has
been associated with the best response in a given How should drug-refractory patients be
individual.70 Once that concentration has been identifi ed managed?
empirically, it might serve as a useful reference to Patients who do not achieve sustained seizure freedom
adjust dosage when a pharmacokinetic change is after adequate trials of at least two appropriate AEDs,
anticipated, such as during pregnancy or when a given alone or in combination, meet ILAE criteria for
potentially interacting drug is added or removed, or to pharmaco resistance as defi ned in a 2010 position paper.79
better assess possible causes of an unexpected change The rationale for this defi nition is that the probability of
in clinical response. The use of serum AED seizure freedom on another AED decreases in proportion
www.thelancet.com/neurology Vol 10 May 2011
Suggested titration rate
Suggested initial
Usual maintenance doses
Frequency of administration
target maintenance (mg per day)
dose (mg per day)
Start with 100 mg per day or 200 mg per day and increase to target Start with 10 mg per day; if indicated, increase to 20 mg per day Start with 400 mg per day and increase to target dose after Start with 250 mg per day and increase to target dose over Start with 600 mg per day or 1200 mg per day and increase to Start with 300–900 mg per day and increase to target dose over Start with 100 mg per day and increase by 100 mg after 1–2 weeks; if indicated, increase further by 100 mg after 1–2 weeks Start with 25 mg per day for 2 weeks, then increase to 50 mg per day for 2 weeks; further increases by 50 mg per day every 1–2 weeks Start with 25 or 50 mg per day for 2 weeks, then increase to 50 or 100 mg per day for 2 weeks; further increases by 50–100 mg per day every 1–2 weeks Start with 25 mg on alternate days for 2 weeks, then 25 mg per day for 2 weeks; further increases by 25–50 mg per day every 1–2 weeks Start with 500 mg per day or 1000 mg per day and increase if Start with 300 mg per day and increase to target dose over Start with 30–50 mg at bedtime and increase, if indicated, after Start with 100 mg per day and increase to target dose over 3–7 days Start with 50–75 mg per day and increase to 150 mg over 2–4 weeks; further increases, if indicated, by increments of 75–150 mg every 2 weeks Start with 62·5 mg per day and increase to target dose over about 3 weeks; in patients on enzyme-inducing comedication, faster titration can be used Start with 200–400 mg per day and increase by 200–400 mg per day after 2 weeks; further increases, if indicated, by 400 mg per day every 2 weeks Start with 5 mg per day and increase by 5 mg increments at weekly Start with 25 mg per day and increase by 25 mg or 50 mg Start with 500 mg per day and increase, if indicated, after about be sometimes feasible, especially with sustained-release formulations) Start with 250 mg per day or 500 mg per day and increase to target 1000 Start with 50 mg per day and increase to 100 mg per day after 1 week; further increases by 50 mg per day every 1–2 weeks or by 100 mg per day after 2 weeks This information refl ects the authors’ experience and might diff er from that reported in product information sheets. Diff erent titration and dosing schemes might be indicated in relation to the clinical context and individual patient characteristics. Modifi ed from Perucca and colleagues,64 by permission of Wolters Kluwer. *Suggested target dose for initial monotherapy in adults with newly diagnosed epilepsy; larger doses might be appropriate for patients with pharmacoresistant epilepsy.
Table 4: Suggested titration rates, initial target maintenance doses, range of maintenance doses, and frequency of administration of antiepileptic drugs in adults
www.thelancet.com/neurology Vol 10 May 2011
to the number of drugs tried unsuccessfully in the past, frequency or severity and the burden of side-eff ects. The and is probably no greater than 20% after failure of two greatest threat for pharmacoresistant patients is use of such drugs.71,75,76 The main objective of the ILAE defi nition excessive AED dosages and polytherapy.95 In these of pharmac oresistance is to emphasise that individuals patients, reduction of overtreatment can provide who did not respond to two AEDs should be referred to a signifi cant benefi ts in terms of reduced adverse eff ects96 specialist service for diagnostic reassessment, including and improved quality of life.97 Addressing comorbidities, assessment of the feasibility of epilepsy surgery. The particularly depression, can be more important than potentially curative eff ect of such surgery is under- focusing on attempts to suppress seizures. Improvement estimated worldwide, as even in developed countries of depressive symptoms might also be achieved with patients are typically referred for surgery 20 years or non-pharmacological interventions.98more after diagnosis.80 Pharmacoresistance is not synonymous with medical Should AEDs be discontinued in seizure-free
intractability, and AED trials are needed for patients patients?
who are not candidates for surgery. Retrospective studies
Discontinuation of AED treatment might be considered suggest that changes in AED treatment could result in after at least 2–4 years of seizure freedom, but only after seizure remission in up to 28% of patients with careful discussion with the patient about associated refractory epilepsy.81,82 In prospective randomised, risks and potential benefi ts. Consideration should be controlled trials, however, seizure freedom rates after given not only to prognostic factors, but also to the addition of a new AED in patients who did not respond presence of adverse eff ects from ongoing medication, to several drugs are less than 10%.83,84 the individual’s lifestyle, and the patient’s attitude How can AED therapy be optimised in pharmacoresistant towards both continuation of treatment and the individuals? Seizure freedom should be pursued, but not possibility of relapse.26,99at all costs. Some patients might benefi t from judicious In general, the risks of seizure recurrence tend to be use of AED combinations, and evidence exists that carefully greater in adults than in children, partly because syndromes individualised polytherapy does not necessarily increase with high rates of spontaneous remission, such as Rolandic the burden of adverse eff ects.77,85 Aspects to be considered epilepsy and childhood absence epilepsy, are typically when trying an additional AED include its spectrum of paediatric.99 The psychosocial consequences of relapse, effi cacy, its adverse eff ects profi le, its expected eff ect based including loss of permission to drive, might also be more on the patient’s characteristics, and the possibility of severe in adults, although the distress associated with pharmacokinetic and pharmacodynamic drug interactions, taking long-term medication should not be under- which can require dosage adjustments.86 Pharmaco- estimated.100 A 2004 review99 of 28 studies comprising dynamic interactions can be especially relevant for drug 4571 patients reported that the cumulative probability of selection, because they might aff ect both effi cacy and remaining seizure free after AED withdrawal was 66–96% tolerability. Although well designed randomised controlled at 1 year and 61–91% at 2 years for children, and 39–74% at trials addressing this topic have not been done, evidence 1 year and 35–57% at 2 years for adults. Predictors of higher from observational studies and from post-hoc analyses of a than average risk of recurrence were adolescent-onset few trials suggests that outcomes of combination therapy epilepsy (which includes juvenile myoclonic epilepsy), can be predicted to some extent by knowledge of the focal seizures, underlying neurological disorder, and for mechanisms of action of individual AEDs. In particular, children, abnormal EEG at the time of withdrawal. Factors experimental87 and clinical78 studies suggest that the associated with lower than average risk were childhood-combination of two sodium-channel-blocking AEDs onset epilepsy, idiopathic generalised epilepsy, and for results in restricted additive effi cacy and potentiation of children, normal EEG. Subsequently, Lossius and CNS side-eff ects, whereas combinations of certain AEDs colleagues101 reported the only double-blind randomised with diff erent mechanisms of action can be advantageous.
controlled trial to date about outcomes after withdrawal. In agreement with this notion, combinations of The study enrolled 160 adults who were seizure free for at lamotrigine with carbamazepine,88 oxcarbazepine with least 2 years, 67% of whom had been seizure free for more carbamazepine,89 rufi namide with carbamazepine,90 or than 5 years (an important consideration because relapse lacosamide with carbamazepine, oxcarbazepine, risk seems to decrease with increasing duration of seizure phenytoin, or lamotrigine91 (all sodium-channel blockers) freedom). At 12 months, relapse rates were 15% among seem to be less advantageous than combinations of the patients randomised to AED withdrawal and 7% among same agents with other drugs. Conversely, combinations those randomised to continue medication. More patients of lamotrigine with valproate92–94 and, possibly, lacosamide in the withdrawal group than in the continuation group with non-sodium-channel-blocking AEDs91 seem to off er showed some improvement in neuropsychological increased benefi t in some patients.
performance, but there were no signifi cant diff erences in When seizure freedom cannot be achieved, the ultimate quality of life. Thus, the risks during the limited follow-up goal is the best possible quality of life, to be obtained period were relatively modest, but the benefi ts were equally through a compromise between reduction in seizure non-impressive.
www.thelancet.com/neurology Vol 10 May 2011
after at least 2–4 years of seizure freedom should be Search strategy and selection criteria
considered, after careful assessment of potential benefi ts versus the risk of relapse and related implications.
References for this Review were identifi ed through searches of PubMed until Feb 28, 2011, with the search terms Contributors
Both authors contributed equally to the preparation of this Review.
“epilepsy”, “treatment”, “antiepileptic drugs”, “effi “eff ectiveness”, “adverse eff ects”, and “humans”. References Confl icts of interest
EP received research funds from the European Union, the Italian
were also identifi ed from relevant review articles and through Ministry of Health, the Italian Ministry for Education and University, searches of the authors’ fi les. Only articles published in and the Italian Medicines Agency. EP has also received speaker’s English were reviewed. The fi nal reference list was based on or consultancy fees and/or research grants from Bial, Eisai, relevance to the topics covered in the Review.
GlaxoSmithKline, Johnson & Johnson, Novartis, Pfi zer, Sanofi -Aventis, SK Holdings, Supernus, UCB Pharma, and Valeant and has been on advisory boards for Bial, Valeant, GlaxoSmithKline, Johnson & Johnson, UCB Pharma, Eisai, Novartis, Pfi zer, and Sanofi -Aventis. TT received An important consideration before discontinuation of research grants and/or speakers honoraria from Eisai, GlaxoSmithKline, AEDs is whether reinstitution of treatment will restore Janssen-Cilag, Novartis, Pfi zer, Sanofi -Aventis, and UCB Pharma.
seizure control in patients who have relapsed. Although a Acknowledgments
large randomised controlled trial provided no evidence This work was partly supported by a grant from the Italian Ministry
that discontinuation modifi es the long-term prognosis of of Health (RC 2009) to the Istituto di Ricovero e Cura a Carattere
a person’s epilepsy,102 a more recent review103 concluded Scientifi co (IRCCS) Fondazione C, Mondino, Pavia, Italy.
that some patients might take more than 1 year to regain References
seizure control, and in 19% of cases (mean of 14 studies) 1
Epilepsy Foundation. Epilepsy and seizure statistics. http://www.
epilepsyfoundation.org/about/statistics.cfm (accessed Dec 31, 2010).
resumption of medication did not control the epilepsy as 2 Hesdorff er DC, Logroscino G, Benn EKT, Katri N, Cascino G, well as before. Factors associated with poor outcome after Hauser WA. Estimating risk for developing epilepsy. A population- a recurrence were symptomatic epilepsy, focal epilepsy, based study in Rochester, Minnesota. Neurology 2011; 76: 23–27.
Kale R. Bringing epilepsy out of the shadows. BMJ 1997; 315: 2–3.
and cognitive defi cits. The authors concluded that “serious International League Against Epilepsy/International Bureau for and substantial risks weigh against discontinuation of Epilepsy/World Health Organization Global Campaign Against AEDs in seizure-free patients, except perhaps for selected Epilepsy. Epilepsy in the WHO European region (2010). http://www.
ibe-epilepsy.org/downloads/EURO%20Report%20160510.pdf patients with idiopathic epilepsy syndromes of childhood Tomson T, Beghi E, Sundqvist A, Johannessen SI. Medical risks Overall, available evidence suggests that the decision to in epilepsy: a review with focus on physical injuries, mortality, withdraw medication needs to be individualised, and that c accidents and their prevention. Epilepsy Res 2004; 60: 1–16.
Sillanpää M, Shinnar S. Long-term mortality in childhood-onset the risks and adverse implications are greater in adults epilepsy. N Engl J Med 2010; 363: 2522–29.
than in children. Although there is insuffi The Lancet Neurology. Epilepsy must become a higher priority to guide the rate of AED discontinuation in adults,104 the in Europe. Lancet Neurol 2010; 9: 941.
process should be gradual and carried out over no less 8 Raspall-Chaure M, Neville BG, Scott RC. The medical management of the epilepsies in children: conceptual and practical than 2–3 months, and preferably longer (at least 6 months) considerations. Lancet Neurol 2008; 7: 57–69.
when withdrawing barbiturates and benzodiazepines.26 Shorvon SD. Drug treatment of epilepsy in the century of the ILAE:
the fi rst 50 years, 1909–1958. Epilepsia 2009; 50 (suppl 3): 69–92.
Conclusions
10 Perucca E, Beghi E, Dulac O, Shorvon S, Tomson T. Assessing risk to benefi t ratio in antiepileptic drug therapy. Epilepsy Res 2000; The key to optimum epilepsy management is to adapt 41: 107–39.
treatment decisions to the characteristics of the individual. 11 Tomson T, Nashef L, Ryvlin P. Sudden unexpected death in epilepsy: current knowledge and future directions. Lancet Neurol In general, AED monotherapy is indicated after two 2008; 7: 1021–31.
seizures, but in high-risk patients initiation of treatment 12 Berg AT, Shinnar S. The risk of seizure recurrence following a fi rst after one seizure might be justifi able. AED choice is unprovoked seizure: a quantitative review. Neurology 1991;
41:
determined by seizure type, adverse-eff ect profi le, and 13 Hauser WA, Rich SS, Lee JR, Annegers JF, Anderson VE. Risk of patient-specifi c features, including age, sex (with special recurrent seizures after two unprovoked seizures. N Engl J Med reference to childbearing potential), and comorbidities. 1998; 338: 429–34.
Dose titration and dosing regimens also need to be 14 Fisher RS, van Emde Boas W, Blume W, et al. Epileptic seizure and epilepsy: defi nitions proposed by the International League against carefully individualised. Failure to achieve seizure freedom Epilepsy (ILAE) and the International Bureau for Epilepsy (IBE). should prompt diagnostic reassessment and early Epilepsia 2005; 46: 470–72.
consideration of the feasibility of epilepsy surgery. 15 Commission on epidemiology and prognosis. International League Against Epilepsy. Guidelines for epidemiological studies on Combination therapy can be benefi cial in patients who did epilepsy. Epilepsia 1993; 34: 592–96.
not respond to two or three sequential monotherapies, 16 First Seizure Trial Group (FIRST Group). Randomized clinical trial although in some cases earlier institution of polytherapy cacy of antiepileptic drugs in reducing the risk of relapse after a fi rst unprovoked tonic clonic seizure. Neurology 1993; might be justifi ed. Seizure freedom is always the ultimate 43: 478–83.
goal, but it should not be pursued at all costs and 17 Marson A, Jacoby A, Johnson A, et al. Immediate versus deferred overtreatment should be carefully avoided. In patients antiepileptic drug treatment for early epilepsy and single seizures: who do achieve seizure control, discontinuation of AEDs a randomised controlled trial. Lancet 2005; 365: 2007–13.
www.thelancet.com/neurology Vol 10 May 2011
18 Musicco M, Beghi E, Solari A, Viani F. Treatment of fi rst 37 Wyszynski DF, Nambisan M, Surve T, et al. Antiepileptic Drug tonic-clonic seizure does not improve the prognosis of epilepsy. Pregnancy Registry. Increased rate of major malformations in Neurology 1997; 49: 991–98.
off spring exposed to valproate during pregnancy. Neurology 2005; 19 Leone MA, Solari A, Beghi E, FIRST Group. Treatment of the fi rst 64: 961–65.
tonic-clonic seizure does not aff ect long-term remission of 38 Morrow J, Russell A, Guthrie E, et al. Malformation risks of epilepsy. Neurology 2006; 67: 2227–29.
antiepileptic drugs in pregnancy: a prospective study from the UK 20 Jacoby A, Gamble C, Doughty J, Marson A, Chadwick D, Medical Epilepsy and Pregnancy Register. J Neurol Neurosurg Psychiatry Research Council MESS Study group. Quality of life outcomes 2006; 77: 193–98.
of immediate or delayed treatment of early epilepsy and single 39 Veiby G, Daltveit AK, Engelsen BA, Gilhus NE. Pregnancy, delivery, seizures. Neurology 2007; 68: 1188–96.
and outcome for the child in maternal epilepsy. Epilepsia 2009; 21 Kim LG, Johnston TL, Marson AG, Chadwick DW, on behalf 50: 2130–39.
of the MRC MESS Study group. Prediction of risk of recurrence 40 Jentink J, Loane MA, Dolk H, et al. Valproic acid monotherapy after a single seizure and early epilepsy: further results from in pregnancy and major congenital malformations. the MESS trial. Lancet Neurol 2006; 5: 317–22.
N Engl J Med 2010; 362: 2185–93.
22 Marson AG. When to start antiepileptic drug treatment and with 41 Meador KJ, Baker GA, Browning N, et al. Cognitive function what evidence? Epilepsia 2008; 49 (suppl 9): 3–6.
at 3 years of age after fetal exposure to antiepileptic drugs. 23 Glauser T, Ben-Menachem E, Bourgeois B, et al. ILAE treatment N Engl J Med 2009; 360: 1597–605.
guidelines: evidence-based analysis of antiepileptic drug effi 42 Samren EB, van Duijn CM, Koch S, et al. Maternal use of and eff ectiveness as initial monotherapy for epileptic seizures antiepileptic drugs and the risk of major congenital malformations: and syndromes. Epilepsia 2006; 47: 1094–120.
a joint European prospective study of human teratogenesis 24 Perucca E, Tomson T. Monotherapy trials with the new associated with maternal epilepsy. Epilepsia 1997; 38: 981–90.
antiepileptic drugs: study designs, practical relevance and ethical 43 Samren EB, van Duijn CM, Christiaens GC, Hofman A, implications. Epilepsy Res 1999; 33: 247–62.
Lindhout D. Antiepileptic drug regimens and major congenital 25 French JA, Kanner AM, Bautista J, et al. Effi abnormalities in the off spring. Ann Neurol 1999; 46: 739–46.
of the new antiepileptic drugs, I: treatment of new-onset epilepsy: 44 Kaneko S, Battino D, Andermann E, et al. Congenital malformations report of the TTA and QSS Subcommittees of the American due to antiepileptic drugs. Epilepsy Res 1999; 33: 145–58.
Academy of Neurology and the American Epilepsy Society. 45 Nicolson A, Appleton RE, Chadwick DW, Smith DF. The Epilepsia 2004; 45: 401–09.
relationship between treatment with valproate, lamotrigine, 26 National Institute for Clinical Excellence. Clinical guideline 20 and topiramate and the prognosis of the idiopathic generalised the epilepsies: the diagnosis and management of the epilepsies epilepsies. J Neurol Neurosurg Psychiatry 2004; 75: 75–79.
in adults and children in primary and secondary care. October 46 Mohanraj R, Brodie MJ. Outcomes of newly diagnosed idiopathic 2004. http://www.nice.org.uk/nicemedia/live/10954/29532/29532.
generalized epilepsy syndromes in a non-pediatric setting. Acta Neurol Scand 2007; 115: 204–08.
27 Scottish Intercollegiate Guidelines Network. Diagnosis and 47 Pennell PB, Peng L, Newport DJ, et al. Lamotrigine in pregnancy. management of epilepsy in adults. A national clinical guideline. Clearance, therapeutic drug monitoring and seizure frequency. April 2003, updated Oct 20, 2005. http://www.sign.ac.uk/pdf/ Neurology 2008; 70: 2130–36.
48 Longo B, Forinash AB, Murphy JA. Levetiracetam use in pregnancy. 28 Brodie MJ, Perucca E, Ryvlin P, Ben-Menachem E, Meencke HJ; Ann Pharmacother 2009; 43: 1692–95.
Levetiracetam Monotherapy Study Group. Comparison of 49 Ohman I, Sabers A, de Flon P, Luef G, Tomson T. Pharmacokinetics levetiracetam and controlled-release carbamazepine in newly of topiramate during pregnancy. Epilepsy Res 2009; 87: 124–29.
diagnosed epilepsy. Neurology 2007; 68: 402–08.
50 Gomer B, Wagner K, Saar J, et al. The infl uence of antiepileptic 29 Marson AG, Al-Kharusi AM, Alwaidh M, et al, on behalf drugs on cognition: a comparison of levetiracetam with topiramate. of the SANAD Study group. The SANAD study of eff ectiveness Epilepsy Behav 2007; 10: 486–94.
of carbamazepine, gabapentin, lamotrigine, oxcarbazepine, 51 Hunt S, Russell A, Smithson WH, et al. Topiramate in pregnancy: or topiramate for treatment of partial epilepsy: preliminary experience from the UK Epilepsy and Pregnancy an unblinded randomised controlled trial. Lancet 2007; Register. Neurology 2008; 71: 272–76.
369: 1000–15.
52 Kothare SV, Valencia I, Khurana DS, Hardison H, Melvin JJ, 30 Marson AG, Al-Kharusi AM, Alwaidh M, et al. The SANAD study cacy and tolerability of zonisamide in juvenile of eff ectiveness of valproate, lamotrigine, or topiramate for myoclonic epilepsy. Epileptic Disord 2004; 6: 267–70.
generalised and unclassifi able epilepsy: an unblinded randomised
controlled trial. Lancet 2007; 369: 1016–26.
53 Brodie MJ, Elder AT, Kwan P. Epilepsy in later life. Lancet Neurol 2009; 8: 1019–30.
31 Perucca E, Alexandre V Jr, Tomson T. Old versus new antiepileptic drugs: the SANAD study. Lancet 2007; 370: 313.
54 Brodie MJ, Overstall PW, Giorgi L. Multicentre, double-blind, randomised comparison between lamotrigine and carbamazepine 32 Perucca E. Introduction to the choice of antiepileptic drugs. In: in elderly patients with newly diagnosed epilepsy. The UK Shorvon SD, Perucca E, Engel J Jr, eds. The treatment of epilepsy. Lamotrigine Elderly Study Group. Epilepsy Res 1999; 37: 81–87.
Oxford, UK: Wiley-Blackwell, 2009: 121–40.
55 Rowan AJ, Ramsay RE, Collins JF, et al. New onset geriatric 33 Harden CL, Meador KJ, Pennell PB, et al. Practice parameter epilepsy: a randomized study of gabapentin, lamotrigine and update: management issues for women with epilepsy—focus on carbamazepine. Neurology 2005; 64: 1868–73.
pregnancy (an evidence-based review): teratogenesis and perinatal outcomes. Report of the Quality Standards Subcommittee and 56 Saetre E, Perucca E, Isojärvi J, Gjerstad L, LAM 40089 Study Group. Therapeutics and Technology Subcommittee of the American An international multicenter randomized double-blind controlled Academy of Neurology and American Epilepsy Society. Epilepsia trial of lamotrigine and sustained-release carbamazepine in the 2009; 50: 1237–46.
treatment of newly diagnosed epilepsy in the elderly. Epilepsia 2007;
48: 1292–302.
34 Artama M, Auvinen A, Raudaskoski T, Isojärvi I, Isojärvi J. Antiepileptic drug use of women with epilepsy and congenital 57 Patsalos PN, Perucca E. Clinically important drug interactions malformations in off spring. Neurology 2005; 264: 1874–78.
in epilepsy: interactions between antiepileptic drugs and other
drugs. Lancet Neurol 2003; 2: 473–81.
35 Vajda FJ, Hitchcock A, Graham J, O’Brien T, Lander C, Eadie M. The Australian Register of Antiepileptic Drugs in Pregnancy: 58 Mintzer S, Mattson RT. Should enzyme-inducing antiepileptic drugs the fi rst 1002 pregnancies. Aust N Z J Obstet Gynecol 2007; be considered fi rst-line agents? Epilepsia 2009; 50 (suppl 8): 42–50.
47: 468–74.
59 Goadsby PJ, Sprenger T. Current practice and future directions 36 Wide K, Winbladh B, Kallen B. Major malformations in infants in the prevention and acute management of migraine. exposed to antiepileptic drugs in utero, with emphasis on Lancet Neurol 2010; 9: 285–98.
carbamazepine and valproic acid: a nation-wide, population-based 60 Kanner A. Depression and epilepsy: a review of multiple facets of register study. Acta Paediatr 2004; 93: 174–76.
their close relation. Neurol Clin 2009; 27: 865–80.
www.thelancet.com/neurology Vol 10 May 2011
61 Hesdorff er DC, Berg AT, Kanner AM. An update on antiepileptic 85 Canevini MP, De Sarro G, Galimberti CA, et al. Relationship drugs and suicide: are there defi nitive answers yet? Epilepsy Currents between adverse eff ects of antiepileptic drugs, number of 2010; 10: 137–45.
coprescribed drugs, and drug load in a large cohort of consecutive 62 Spina E, Perugi G. Antiepileptic drugs: indications other than patients with drug-refractory epilepsy. Epilepsia 2010; 51: 797–804.
epilepsy. Epileptic Disord 2004; 6: 57–75.
86 Patsalos PN, Perucca E. Clinically important drug interactions 63 Mula M, Pini S, Cassano GB. The role of anticonvulsant in epilepsy: general features and interactions between antiepileptic drugs in anxiety disorders: a critical review of the evidence. drugs. Lancet Neurol 2003; 2: 347–56.
J Clin Psychopharmacol 2007; 27: 263–72.
87 Deckers CL, Czuczwar SJ, Hekster YA, et al. Selection of 64 Perucca E, Dulac O, Shorvon S, Tomson T. Harnessing the clinical antiepileptic drug polytherapy based on mechanisms of action: potential of antiepileptic drug therapy: dosage optimisation. the evidence reviewed. Epilepsia 2000; 41: 1364–74.
CNS Drugs 2001; 15: 609–21.
88 Besag FM, Berry DJ, Pool F, Newbery JE, Subel B. Carbamazepine 65 Kale R, Perucca E. Revisiting phenobarbital for epilepsy. BMJ 2004; toxicity with lamotrigine: pharmacokinetic or pharmacodynamic 329: 1199–200.
interaction? Epilepsia 1998; 39: 183–87.
66 Perucca E. Extended-release formulations of antiepileptic drugs: 89 Barcs G, Walker EB, Elger CE, et al. Oxcarbazepine rationale and comparative value. Epilepsy Curr 2009; 9: 153–57.
placebo-controlled, dose-ranging trial in refractory partial epilepsy. 67 Gidal BE, Tomson T. Debate: substitution of generic drugs Epilepsia 2000; 41: 1597–607.
in epilepsy: is there cause for concern? Epilepsia 2008; 90 Brodie MJ, Rosenfeld WE, Vazquez B, et al. Rufi namide for the 49 (suppl 9): 56–62.
adjunctive treatment of partial seizures in adults and adolescents: 68 Kesselheim AS, Stedman MR, Bubrick EJ, et al. Seizure outcomes a randomized placebo-controlled trial. Epilepsia 2009; 50: 1899–909.
following the use of generic versus brand-name antiepileptic drugs: 91 Sake JK, Hebert D, Isojärvi J, et al. A pooled analysis of lacosamide a systematic review and meta-analysis. Drugs 2010; 70: 605–21.
clinical trial data grouped by mechanism of action of concomitant 69 Kwan P, Brodie MJ. Eff ectiveness of fi rst antiepileptic drug. antiepileptic drugs. CNS Drugs 2010; 24: 1055–68.
Epilepsia 2001; 42: 1255–60.
92 Brodie MJ, Yuen AW. Lamotrigine substitution study: evidence for 70 Patsalos PN, Berry DJ, Bourgeois BF, et al. Antiepileptic drugs— synergism with sodium valproate? 105 Study Group. Epilepsy Res best practice guidelines for therapeutic drug monitoring: a position 1997; 26: 423–32.
paper by the subcommission on therapeutic drug monitoring, ILAE 93 Wong IC, Mawer GE, Sander JW, Lhatoo SD. A Commission on Therapeutic Strategies. Epilepsia 2008; 49: 1239–76.
pharmacoepidemiologic study of factors infl uencing the outcome 71 Kwan P, Brodie MJ. Early identifi cation of refractory epilepsy. of treatment with lamotrigine in chronic epilepsy. Epilepsia 2001; N Engl J Med 2000; 342: 314–19.
42: 1354–58.
72 Hirsch LJ, Arif H, Nahm EA, Buchsbaum R, Resor SR Jr, Bazil CW. 94 Pisani F, Oteri G, Russo MF, Di Perri R, Perucca E, Richens A. Cross-sensitivity of skin rashes with antiepileptic drug use. cacy of valproate-lamotrigine comedication in refractory Neurology 2008; 71: 1527–34.
complex partial seizures: evidence for a pharmacodynamic
interaction. Epilepsia 1999; 40: 1141–46.
73 Leach JP, Lauder R, Nicolson A, Smith DF. Epilepsy in the UK: misdiagnosis, mistreatment, and undertreatment? The Wrexham 95 Perucca E, Kwan P. Overtreatment in epilepsy: how it occurs area epilepsy project. Seizure 2005; 14: 514–20.
and how it can be avoided. CNS Drugs 2005; 19: 897–908.
74 Faught E, Duh MS, Weiner JR, Guérin A, Cunnington MC. 96 Gilliam FG, Fessler AJ, Baker G, Vahle V, Carter J, Attarian H. Nonadherence to antiepileptic drugs and increased mortality: Systematic screening allows reduction of adverse antiepileptic drug fi ndings from the RANSOM Study. Neurology 2008; 71: 1572–78.
eff ects: a randomized trial. Neurology 2004; 62: 23–27.
75 Perucca E. Pharmacoresistance in epilepsy. How should it be 97 Uijl SG, Adjustment of treatment increases quality of life in defi ned? CNS Drugs 1998; 10: 171–79.
patients with epilepsy: a randomized controlled pragmatic trial.
Eur J Neurol 2009; 16: 1173–77.
76 Schiller Y, Najjar Y. Quantifying the response to antiepileptic drugs: eff ect of past treatment history. Neurology 2008; 70: 54–65.
98 Ciechanowski P, Chaytor N, Miller J, et al. PEARLS depression treatment for individuals with epilepsy: a randomized controlled 77 Beghi E, Gatti G, Tonini C, et al. Adjunctive therapy versus trial. Epilepsy Behav 2010; 19: 225–31.
alternative monotherapy in patients with partial epilepsy failing on a single drug: a multicentre, randomised, pragmatic controlled 99 Specchio LM, Beghi E. Should antiepileptic drugs be withdrawn trial. Epilepsy Res 2003; 57: 1–13.
in seizure-free patients? CNS Drugs 2004; 18: 201–12.
78 Kwan P, Brodie MJ. Combination therapy in epilepsy: when and 100 Jacoby A, Johnson A, Chadwick D. Psychosocial outcomes of what to use. Drugs 2006; 66: 1817–29.
antiepileptic drug discontinuation. The Medical Research Council Antiepileptic Drug Withdrawal Study Group. Epilepsia 1992; 79 Kwan P, Arzimanoglou A, Berg AT, et al. Defi nition of drug 33: 1123–31.
resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies. Epilepsia 2010; 101 Lossius MI, Hessen E, Mowinckel P, et al. Consequences of 51: 1069–77.
antiepileptic drug withdrawal: a randomized, double-blind study
(Akershus study). Epilepsia 2008; 49: 455–63.
80 Spencer SS, Berg AT, Vickrey BG, et al. Initial outcomes in the Multicenter Study of Epilepsy Surgery. Neurology 2003; 61: 1680–85.
102 Chadwick D, Taylor J, Johnson T. Outcomes after seizure recurrence in people with well-controlled epilepsy and the factors that 81 Callaghan BC, Anand K, Hesdorff er D, Hauser WA, French JA. infl uence it. The MRC Antiepileptic Drug Withdrawal Group. Likelihood of seizure remission in an adult population with Epilepsia 1996; 37: 1043–50.
refractory epilepsy. Ann Neurol 2007; 62: 382–89.
103 Schmidt D, Löscher W. Uncontrolled epilepsy following 82 Luciano AL, Shorvon SD. Results of treatment changes in patients discontinuation of antiepileptic drugs in seizure-free patients: with apparently drug-resistant chronic epilepsy. Ann Neurol 2007; a review of current clinical experience. Acta Neurol Scand. 2005; 62: 375–81.
111: 291–300.
83 Beyenburg S, Stavem K, Schmidt D. Placebo-corrected effi 104 Sirven JI, Sperling M, Wingerchuk DM. Early versus late of modern antiepileptic drugs for refractory epilepsy: systematic antiepileptic drug withdrawal for people with epilepsy in remission. review and meta-analysis. Epilepsia 2010; 51: 7–26.
Cochrane Database Syst Rev 2001; 3: CD001902.
84 Rheims S, Perucca E, Cucherat M, Ryvlin P. Factors determining response to antiepileptic drugs in randomized controlled trials.
A systematic review and meta-analysis. Epilepsia 2011; 52: 219–33.
www.thelancet.com/neurology Vol 10 May 2011

Source: http://cluster011.ovh.net/~neurobre/joomla1.5/images/stories/Lancet2011ThePharmacologicalTreatmentofEpilepsy_in_adults.pdf