Epilepsia, 45(Suppl.

6):2834, 2004
Blackwell Publishing, Inc.

C International League Against Epilepsy

The Use of Antiepileptic DrugsPrinciples and Practice

Josemir W. Sander
Department of Clinical and Experimental Epilepsy, Institute of Neurology, University College London, and National Society for
Epilepsy, Chasfort St. Peter, U.K.

Summary: Up to 70% of people developing epilepsy may

expect to become seizure free with optimum antiepileptic drug
(AED) therapy. The remaining 30% are the most difficult to
treat. Most patients are controlled on a single AED, but a
small proportion requires a combination of two agents. Add-on
therapy with a second drug, rather than substitution, may be a
viable and rational approach in some patients, particularly if the
first drug is relatively well tolerated. Precise classification of
the type of seizures, as well as the epilepsy syndrome, together
with careful recording of both seizures and adverse effects,
are essential if rational management decisions are to be made.
The goal of therapy should be complete seizure freedom with
a single drug taken once or twice a day and without adverse
effects. If control is difficult to achieve, the maximum tolerated
dose of each drug should be explored, but a balance needs

to be struck between adverse effects and control of seizures.

In patients in whom treatment appears to be ineffective,
the diagnosis of epilepsy and adherence to therapy should
be reviewed. Drugs used in combination must be carefully
selected, as poor adherence, drug interactions, and toxicity are
more likely if more than one drug is prescribed. Agents are
usually chosen according to seizure type, patient characteristics,
and often by clinician preference. Those that are better tolerated
have a low potential for pharmacokinetic and pharmacodynamic
interactions, and those that can be easily introduced without any
complicated titration schedule have an advantage. Key Words:
EpilepsyAntiepileptic drugsTreatment principlesPatient factorsDrug factorsCombination therapyReal-life
practice.

Epilepsy is among the most common serious neurological conditions. The age-adjusted incidence of epilepsy in
developed countries is around 50 per 100 000 persons per
year (range, 2470 per 100 000 persons per year) (1) and
the prevalence is between 4 and 10 per 1,000 persons (1).
The incidence of epilepsy is higher in infants and the elderly than in those of intermediate age, and is lower in
higher socioeconomic classes (1).
Nonpharmacological management options [e.g.,
surgery (2), vagus nerve stimulation (2)] are feasible
approaches in a few selected individuals with epilepsy. In
the rare cases where clear precipitants to seizures can be
identified, avoidance therapy may be indicated (3).
Given the relatively few patients who can be managed
with nonpharmacological options, antiepileptic drugs
(AEDs) are the mainstay of treatment. The primary goals
of AED treatment are to achieve complete seizure freedom, ideally without adverse events (1,4), reduce morbidity and mortality, and improve quality of life (5). Optimum
AED therapy can abolish seizures in up to 70% of patients
developing epilepsy, but in the remaining patients remission is elusive (1,6).

Antiepileptic drugs are selected first and foremost according to clinical efficacy, then tolerability, drug interaction profile, and ease of use (7). Most patients are controlled on monotherapy (7): 67 and 69% of 421 patients
receiving monotherapy responded to an established or a
newer AED, respectively, in a prospective Scottish study
(6). Despite the good response to monotherapy in many
patients, up to 50% (7,8) will be managed with combination therapy.
Monotherapy is currently the best pharmacotherapeutic
option and is mandatory when first starting AED treatment
(9). An appropriate drug for the seizure type/syndrome
with low adverse event risk and expected best efficacy
should be selected, using an appropriate dosage and slowly
titrating to response (8,10).
If monotherapy is poorly tolerated or ineffective, the
strategy is to switch to another drug. If the first drug has
partial efficacy and is well tolerated, it is worth trying
another drug in combination (11). Add-on therapy appears to be more effective when started immediately after first-drug failure rather than after a second drug has
also failed (12). Outcome among patients stopping therapy with the first drug was strongly correlated with the
reason for treatment discontinuation (p < 0.001) in an
observational study, with four to five times more patients
responding to a subsequent drug or drug combination if

Address correspondence and reprint requests to Dr. J. W. Sander at

Department of Clinical and Experimental Epilepsy, Institute of Neurology, University College London, Queen Square, London WC1N 3BG,
U.K. E-mail: lsander@ion.ucl.ac.uk

28

PRINCIPLES AND PRACTICE OF ANTIEPILEPTIC DRUG USE

FIG. 1. Treatment pathway for newly diagnosed epilepsy [modified from Brodie and Kwan (7) and Kwan and Brodie (12)].

treatment failure with the first drug was due to adverse

events/idiosyncratic reactions rather than lack of efficacy
(6). Fig. 1 shows a treatment pathway for people with
newly diagnosed epilepsy.
Whether the patient is treated with monotherapy or
combination therapy, management plans need to be tailored to the individual, considering not just the epilepsy
type/syndrome but also AED factors and patient characteristics/preferences (4,7). This article reviews the principles of managing epilepsy in this context and provides a
commentary on real-life clinical practice.
THE IMPACT OF SEIZURE TYPE/SYNDROME
ON TREATMENT SELECTION
Correct diagnosisconfirming that the seizures are true
epileptic seizuresand classification of the seizure type
or syndrome are the starting point for successful pharmacotherapy (1,4,11,13,14). Diagnosis is made by taking
a patient history and/or witness account of the seizure, if
available (1). Syncope (14) (the most common differential
diagnosis), other cerebral disorders and metabolic impairments (11) should be ruled out before starting therapy.
In reality, false positive and false negative diagnoses are
frequent (1,14).
Clinical characteristics and EEG patterns, where available, can establish whether the seizure type is generalized
(e.g., absence, myoclonic, tonic-clonic, atonic) or partial
(focal) simple or complex seizures with or without secondary generalization (11,13).
Defining the epilepsy syndrome takes into account
seizure type, presence of structural lesions, age at onset, family history, and EEG changes (14). Syndromes
are broadly categorized as idiopathic (genetic), symptomatic (structural abnormality identified), or cryptogenic
(anatomical abnormality presumed but not demonstrated)
(11,15).
Prognosis is generally excellent in patients with benign partial epilepsies and those with seizures triggered
by specific precipitants. It is good in childhood absence

29

seizures, epilepsy with nonspecific generalized tonicclonic seizures, and some localization-related epilepsies.
These seizure types usually respond well to treatment and
remission may be spontaneous (14).
Prognosis is uncertain in juvenile myoclonic epilepsy
and many localization-related epilepsies; relapse is common when AED treatment is withdrawn. Prognosis is
poor in disorders associated with congenital neurological deficits, progressive neurological disorders, and some
symptomatic or cryptogenic partial epilepsies unresponsive to aggressive treatment (14).
Defining refractory epilepsy is difficult but the condition appears to be multifactorial and progressive, and can
substantially worsen quality of life (4). Patients presenting with many seizures initially or who respond poorly
to initial AED treatment may have refractory disease at
the outset (6). In a prospective community-based cohort
study of nearly 750 patients in the U.K. (16), the most important factor predicting outcome was number of seizures
occurring within the first 6 months after presentation.
The decision to treat, then, relies on weighing the risk
of not treating a specific condition against the benefits
of treatment/likelihood of adverse events with specific
drugs. Distinguishing one type of seizure with good expected outcome (e.g., absence seizures) from another with
similar clinical presentation but a poorer prognosis (e.g.,
complex partial seizures (13)) is crucial when selecting
AEDs. Treatment is warranted in patients with generalized tonic-clonic, tonic, and atonic seizures in whom the
mortality/morbidity risk is high (14).
IMPACT OF DRUG CHARACTERISTICS ON
TREATMENT SELECTION
Efficacy
In epilepsy management, rational AED selection is
based primarily on the spectrum of efficacy and adverse
events profile (Table 1) (7,10,15); efficacy is the most important drug factor (7). In everyday practice, though, the
clinicians comfort level with a drug (7) and the ability to
optimize dosage and co-medications (ease of use) (17) can
have more than just a passing influence on decision making. Drugs with once or twice daily dosing regimens and
uncomplicated titration schedules have an advantage.
Established AEDs (Table 1) have a broad spectrum
of activity against partial and tonic-clonic seizures with
or without secondary generalization (10,15). Although
demonstrable differences in efficacy are absent among
established AEDs (5,7), some newer drugs may be better choices than others in patients with comorbid conditions such as bipolar disorders, migraine (5), and sleep
disturbance.
Monotherapy is the target, but in reality as many as 50%
of patients are managed with two drugs or more (18). In a
Scottish epilepsy unit database, 21% of 1,617 seizure-free
Epilepsia, Vol. 45, Suppl. 6, 2004

30

J. W. SANDER

TABLE 1. Main indications, advantages and disadvantages of established and newer antiepileptic drug [adapted from Brodie and
Kwan (7) and Perucca (10); additional data from Foldvary (15)]
Drug
Established drugs
Carbamazepine

Main indications

Advantages

Partial and tonic-clonic

Known mechanism
Modified-release formulation

Clobazam

Partial and tonic-clonic

Phenobarbital

Partial, tonic-clonic, and clonic

Phenytoin

Partial, tonic-clonic, and status

epilepticus

Lacks neuropsychiatric effects

No drug interactions
Long half-life
Known mechanism
Known mechanism
Lacks neuropsychiatric effects

Valproate

Generalized and partial seizures

including absence, myoclonic, and
tonic-clonic

Lacks neuropsychiatric effects

Severe and/or refractory epilepsies

(Lennox-Gastaut syndrome)

Broad spectrum

Gabapentin

Adjunctive treatment of refractory

partial seizures with/without
secondary generalization

Lamotrigine

Adjunctive therapy and monotherapy

of partial and generalized seizures

Levetiracetam

Adjunctive treatment of refractory

partial seizures with/without
secondary generalization

Oxcarbazepine

Adjunctive treatment and

monotherapy of partial seizures and
primary or secondary generalized
seizures

Tiagabine

Adjunctive treatment of refractory

partial seizures with/without
secondary generalization

Well tolerated
No idiosyncratic reactions
No drug interactions
Rapid titration
Effective against some comorbidities
(neuropathic pain)
Good CNS profile
Twice daily dosing
Broad spectrum
Predictable kinetics
Additive reaction with valproate
Efficacy against some comorbidities
(e.g., bipolar depression)
Well tolerated
Twice daily dosing
Rapid titration
No drug interactions
Well tolerated (including CNS profile)
Ease of titration
Less allergenic than carbamazepine
Efficacy against some comorbidities
(acute mania)
Known mechanism
Lacks idiosyncratic reactions
Not an enzyme inducer
Low allergenic potential

Topiramate

Adjunctive treatment of partial and

generalized seizures

Twice daily dosing

Low allergenic potential
High efficacy
No idiosyncratic reactions
Efficacy against some comorbidities
(binge eating disorder)

Vigabatrin

Adjunctive treatment of partial

seizures with/without secondary
generalization refractory to all
other drugs; infantile spasms (West
syndrome)
Adjunctive treatment of refractory
partial and generalized seizures

Ease of titration
No important drug interactions
Good efficacy in infantile spasms

Newer drugs
Felbamate

Zonisamide

Epilepsia, Vol. 45, Suppl. 6, 2004

Twice daily dosing

Disadvantages
Autoinduction
Enzyme inducer
Neurotoxic
Idiosyncratic reactions (aplastic anemia,
hyponatremia, impaired renal function)
Teratogenic
Sedative
Tachyphylaxis
Sedative and other CNS effects
Idiosyncratic reactions
Cosmetic adverse effects (hirsutism, acne,
gingival hyperplasia)
Enzyme inducer
Teratogenic
Enzyme inducer
High protein binding
Can cause weight gain, idiosyncratic reactions
and, rarely, hepatic failure and fatal
pancreatitis
Teratogenicity
Unknown mechanism
Risk of severe adverse effects (aplastic
anemia, hepatic failure)
CNS and GI effects
Drug interactions
Requires close monitoring
for adverse effects
Variable absorption
Weight gain
Three times daily dosing
Limited spectrum of activity
Slow titration
Interaction with carbamazepine and other
enzyme inducers/inhibitors
Dose-related rash
May aggravate severe myoclonic epilepsy
Data lacking in generalized epilepsies

Limited spectrum
Interacts with oral contraceptives
Rash, hyponatremia
Teratogenic potential
Limited spectrum
Slow titration
Three times daily dosing
Inducible metabolism
Dizziness
Slow titration
CNS adverse effects
Interacts with oral contraceptives
Renal stones
Teratogenic potential
Efficacy against absence seizures not
established
Visual field defects
Limited spectrum
Weight gain
Slow titration
CNS adverse effects and allergic reactions

PRINCIPLES AND PRACTICE OF ANTIEPILEPTIC DRUG USE

Adverse effects
Given the apparent lack of differences in efficacy among
AEDs, tolerability is a key consideration when selecting
AED therapy (5,10). Although complete seizure freedom
is generally thought to be the most significant predictor
of improved quality of life (5), adverse effects may be the
most important negative influence on a persons perception of individual current health status (20). Significant adverse events associated with established AEDs contribute
to initial treatment failure in >40% of patients (20).
The distinct differences in the type, although not necessarily the incidence (5), of unwanted effects among various
AEDs can be exploited when selecting therapy. Knowing,
for example, that valproate (VPA) may cause weight gain
and topiramate (TPM) may cause weight loss could sway
decision making (5) when selecting options for under- or
overweight individuals. At least in theory, it should be possible to combine AEDs with similar spectrum of activity
but different adverse event profiles to avoid overlap and
minimize unwanted effects (8).
Flexible titration to response and controlled dose escalation rate improves the tolerability of many drugs (10).
If achieving seizure control is difficult, the maximum tolerated dose of the drug should be explored, but a balance
needs to be struck between adverse effects and seizure
control (5). Careful monitoring of seizures and adverse
events is essential if rational management decisions are to
be made.
Serious or potentially life-threatening toxicity restricts
the use of some otherwise highly efficacious AEDs. The
best-known examples are irreversible visual field defects
with vigabatrin, and bone marrow and liver toxicity with
felbamate (5,7,10). If either drug is selected, patients must
be carefully monitored during therapy (10). Valproate
has been associated with infrequent but occasionally fatal
hepatotoxicity, while carbamazepine (CBZ) has a wellpublicized, albeit relatively low, risk of bone marrow suppression (5).
Established AEDs (Table 1), particularly sodium valproate, carry a risk of teratogenicity that increases with
increasing dosages and number of concomitant drugs (7).
Whether the lack of fetal abnormalities seen in animal
models with some newer drugs (7) will translate into reduced risk in humans is being assessed through collaborative registries in pregnant women (10).

Drug interactions and pharmacokinetic

considerations
Pharmacodynamic rather than pharmacokinetic drug
interactions may explain the greater toxicity of some AED
combinations [lamotrigine (LTG) + CBZ] (21), and improved efficacy of others (LTG + VPA) (22) in patients unresponsive to either drug alone. Other combinations may
also benefit from pharmacodynamic interactions [VPA +
ethosuximide, clonazepam or CBZ; LTG + TPM; CBZ +
tiagabine; phenobarbital + phenytoin (PHT)] (23,24) but
for some of these a pharmacokinetic interaction may have
contributed (23).
Pharmacokinetic drug interactions complicate management in patients on combination therapy or those receiving
treatment for comorbid conditions, although dose adjustment and serum drug level monitoring can overcome these
problems (5). Toxicity may be increased and the efficacy
of either the AEDs or the concomitant drugs reduced. Hepatic enzyme-inducing AEDs (e.g., CBZ, PHT) have a high
propensity for drug interactions (8,23). On the other hand,
drugdrug interactions are unlikely with AEDs not metabolized hepatically, notably gabapentin, levetiracetam, and
the new drug pregabalin (25).
Hepatic enzyme-inducing AEDs also reduce the efficacy of oral contraceptives in women with epilepsy
(11,23). Alarmingly, in a U.S. survey of 1,000 neurologists
and 1,000 obstetricians/gynecologists, the responders (n
= 160 and 147, respectively) lacked knowledge and expertise in identifying and managing drug interactions between AEDs and oral contraceptives (26) (Fig. 2). Only
4% of neurologists and no obstetricians correctly identified the effects of the six most commonly used AEDs on
oral contraceptives. A retrospective search of a U.K. prescription database supports these worrying results: 56%
of 200 women taking AEDs and oral contraceptives with
low estrogen content may have been at increased risk
of contraceptive failure (27). This underscores the need

Neurologists (n=160)

Obstetricians (n=147)

30
Percent of physicians

patients were taking more than one drug, and 13% of those
received three (19). It has been suggested that a rational approach might be to combine AEDs with different mechanisms, such as a sodium channel blocker with an agent that
enhances GABAergic inhibition (7). At present, though,
the lack of specific and reliable information on seizure
pathophysiology means this method cannot be adopted
unreservedly (10).

31

25
20
15
10
5
0
Reported OC failure in
patients taking AEDs

Knew effects of 6 most

common AEDs on OCs

FIG. 2. Lack of knowledge among responding neurologists and

obstetricians in a U.S. survey about drug interactions between
antiepileptic drugs (AEDs) and oral contraceptives (OCs) (26)

Epilepsia, Vol. 45, Suppl. 6, 2004

32

J. W. SANDER
TABLE 2. Properties of an ideal antiepileptic drug

High oral efficacy without seizure aggravation

Good tolerability and no teratogenicity
No or minimal drug interactions
Once or twice daily dosing
Range of formulations available
Low cost and high cost effectiveness

for physicians to understand and update their knowledge

of potentially life-altering drug interactions in women of
childbearing potential.
Hepatic enzyme-inducing AEDs may also affect the
response to other drugs metabolized by similar pathways, e.g., warfarin, some tricyclic antidepressants, corticosteroids, some antineoplastic drugs, many antivirals,
and cyclosporin (10,23). Conversely, fluoxetine, cimetidine, omeprazole, fluconazole, and erythromycin may
increase the toxicity of hepatic enzyme-inducing AEDs
(10,23). A second-generation drug not metabolized by the
cytochrome P450 3A system should be selected for patients receiving these drugs (10,23).
Special populations have specific pharmacokinetic considerations. The elderly metabolize and eliminate drugs
more slowly than younger adults; they are likely to be
taking multiple medications for comorbid conditions and
are at increased risk of toxicity, whereas drug metabolism
is more rapid in children (23). The elimination of drugs
cleared primarily via renal mechanisms is decreased in patients with renal failure, and dosage adjustments may be
necessary (5,28). The presence of severe hepatic impairment decreases the serum protein binding and elimination
of some hepatically metabolized drugs (28).
The ideal AED would possess the wish list of properties shown in Table 2 high efficacy, good tolerability, minimal drug interactions, and simple administration
schedules, as well as high cost effectiveness (29). These
properties are especially relevant when managing patients
with combination therapy.
IMPACT OF PATIENT FACTORS ON
TREATMENT SELECTION
The ultimate aim of AED treatment and seizure control is the maintenance of the patients normal lifestyle
and the reduction of epilepsy-induced mortality/morbidity
(5,11). Treating epilepsy involves managing the patient
and lifestyle as a whole, not just controlling the epileptic seizures. Therefore, clinicians need to gain the trust
of the patient, as well as the patients involvement in,
and commitment to, an agreed management plan. Seizure
precipitants should be identified (e.g., sleep deprivation,
exercise, excessive alcohol, missed doses of AEDs, concomitant drugs, and metabolic factors such as hyper- or
hyponatremia) (3) and the patient counseled on how to
avoid these scenarios where possible.
Epilepsia, Vol. 45, Suppl. 6, 2004

Problems in clinical practice among general practitioners, as revealed in a postal questionnaire in the U.K.,
highlight a lack of confidence about their knowledge of
epilepsy, unfamiliarity with new drugs, and little time to
counsel and assess patients (30). To optimize compliance
and treatment success, it is in the best interests of both
patient and clinician to make time to discuss the potential
impact of both the disease (lifestyle limitations and social stigma) and the proposed therapy (potential adverse
events, drug interactions, life-long treatment, importance
of compliance, and implications for contraception and
pregnancy) (11,31). Patients with epilepsy commonly list
fear of next seizure, social stigma, lifestyle limitations
(e.g., driving) and some aspects of having to take drugs
as the worst part of their disease (32). Patient factors such
as age, gender, learning disability, and compliance all affect AED treatment selection and success.
Age
Children and adolescents with epilepsy have a poorer
quality of life than children with other life-long conditions
(33); compared with children with asthma, for example,
children with epilepsy have a more compromised quality
of life in the psychological, social, and school domains
(34). The impact on the child or teenagers self esteem and
on the family unit [particularly for parents who may feel
shame or be overly protective of the child (33)], should not
be underestimated. Elderly patients lose self-confidence,
are more prone to injury after a seizure, and may become
increasingly isolated as functional independence diminishes, e.g., if they are unable to drive (33). Generalized
seizures are more common than partial seizures in children under 5 years old, while the converse is true in the
elderly (35).
Adverse effects of AEDs can also reduce quality of life
in children and the elderly, although for different reasons
(33). Some drugs exacerbate cognitive impairment and
hyperactivity in children (5). Because elderly patients are
more likely to have comorbid conditions requiring combination therapy, drug interactions may be more frequent
and adherence poorer. Antiepileptic drugs can also aggravate age-related cognitive deficits (33). In the elderly,
drugs should be started at low doses to minimize adverse
effects and drug interactions (11).
On the other hand, appropriately selected AEDs started
early enough in the course of the disease reduce the
high risk of cognitive impairment with epileptic encephalopathies (West syndrome, Lennox-Gastaut syndrome, and myoclonic-astatic epilepsy) (5). Vigabatrin,
for example, is highly effective in West syndrome, although the serious visual defects associated with this drug
are difficult to assess in infants (5).
Gender
Women of childbearing age pose particular management challenges. Criticisms of care (both primary and

PRINCIPLES AND PRACTICE OF ANTIEPILEPTIC DRUG USE

secondary) by women surveyed in a focus group discussion included poor provision of information about adverse
effects, drug interactions with oral contraceptives, and perceived haphazard prescribing of AEDs resulting in serious
adverse effects (36).
The risk of unwanted or unplanned pregnancy rises
in women taking oral contraceptives concomitantly with
hepatic enzyme-inducing AEDs (23). In these women, oral
contraceptives containing more than 30 mcg ethinylestradiol should be prescribed, or a noninducing AED substituted if appropriate (11,31).
In pregnant women, the potential danger posed by uncontrolled seizures to both mother and child must be balanced against the risks to the fetus of AED-related major
or minor abnormalities and impaired growth (5). Planning
the pregnancy is crucial and seizures should be well controlled before conception (31). Monotherapy in the lowest effective dose should be attempted wherever possible
(5,11), as combination therapy markedly increases the teratogenic risk (7,11).
Learning disability
More than one in five patients with epilepsy have learning or intellectual disabilities (5), often in tandem with
behavioral disorders (11). To avoid further impairing cognition and worsening behavior, unnecessary polytherapy
should be avoided (5). Paradoxical seizure aggravation
has been reported with such drugs as CBZ and benzodiazepines in specific seizures types, and close monitoring
is essential during therapy with these AEDs (5).
Adherence to drug regimen
The importance of regular AED usage to avoid loss of
seizure control must be stressed before starting therapy
and the patients likely adherence to medication regimens
assessed. In patients not responding to treatment, the diagnosis of epilepsy and adherence to therapy should be
reviewed.
Up to half of patients take their medication differently
from their prescribed regimen (3), either by adjusting the
dosage or the dosing schedule or both (37). In a 10-item
postal survey of epilepsy patients, 71% admitted dose
omissions and 45% reported a seizure after missing a
dose. Seizures after missed doses were associated with
the number of medication doses (p = 0.04) and number
of medication tablets/capsules (p = 0.01) (38), confirming the general belief that once or twice daily dosing is
preferable (28).
Pregnant women with epilepsy frequently stop or reduce their prescribed medication, usually without informing their clinician (39) and without understanding or acknowledging the potential damage to the fetus if maternal
seizures are uncontrolled (31). Selecting a drug with the
fewest possible adverse effects and a convenient dosing
regimen, together with thorough patient advice about the

33

potential consequences of not taking the medication, is

paramount to achieving a good outcome.
DISCUSSION
Seizure characteristics, and drug and patient factors are
all pieces of the puzzle that need to fit together when implementing management strategies in epilepsy. Correct diagnosis and classification is the key initial step. However,
the present system of classification (40) does not fully incorporate recent findings using magnetic resonance imaging (MRI) and genetic research (2). A new classification
system recognizing these new data and techniques, and
distinguishing more clearly between seizures, syndromes,
and diseases, is being developed (2), and it is hoped that
this will define seizure types more precisely. This is particularly important because a good outcome relies in part
on differentiating between seizure types and syndromes
and selecting the most appropriate agent for that type (1).
After the decision is made to start treatment, AED therapy should be no more than a minor inconvenience to the
patient, with few if any adverse events and drug interactions. Given the lack of documented differences in efficacy
between AEDs (5), drugs that are better tolerated, have a
low potential for unwanted pharmacokinetic and pharmacodynamic interactions, and encourage compliance with
simple dosing regimens and titration schedules have an
advantage.
Around 70% of patients become seizure free with good
management (1). Although monotherapy is the goal, 30
40% will need combination therapy to control seizures
(7). When monotherapy fails, add-on therapy with a second drug, rather than substitution, may be a viable and
rational approach, particularly if the first drug is relatively
well tolerated. The finding that outcome is better when a
second drug is added immediately after the first drug fails,
rather than waiting to see whether the second drug works
(12), suggests this strategy has merit for some patients. A
review of the literature found no conclusive evidence for
choosing between alternative monotherapy and switching
to combination therapy when first-line monotherapy fails
(24), but recommended lowering the dose of the first drug
if seizure control was not achieved at the maximum tolerated dose and adding a second drug.
The common perception that combination therapy carries a higher risk of adverse effects also bears further
scrutiny, in light of some evidence that toxicity is similar
between monotherapy and polytherapy when total drug
load is equal (24). Clinicians should be encouraged to
adopt a flexible approach that includes implementing combination therapy in appropriate dosages.
Although principles of epilepsy treatment are relatively
well established (Table 3), and much is now known about
the newer generation and classes of AEDs, real-life practice continues to differ from ideal scenarios. Clinicians
often choose therapy according to their own comfort level
Epilepsia, Vol. 45, Suppl. 6, 2004

34

J. W. SANDER
TABLE 3. Principles of epilepsy management

Confirm diagnosis of true seizure

Evaluate need for treatment initiation
Attempt to establish seizure type and syndrome
Select AED on basis of seizure type, spectrum of activity,
tolerability, and drug interactions
Tailor AED to individual, and individualize dosage based
on response
If seizures persist, either
add another drug if the first one is well tolerated and partly
effective
or
switch to another drug if the first drug is poorly tolerated
or ineffective

14.
15.
16.
17.
18.
19.
20.

with particular AEDs, rather than weighing up all disadvantages and advantages of older versus newer, less familiar drugs (7). Practitioners also perceive themselves, and
are perceived by patients, as lacking information about potentially serious drug interactions or adverse effects. This
is particularly evident for women of childbearing age, who
may take unnecessary risks with their health or that of
their unborn child if consequences are not well explained
(26,27,36,39).
There is room, then, for improvement in practitioner
knowledge, but also for developing better treatment options and strategies and more rational drug selection based
on patient-related predictors of outcome. With the increased availability of easier to use formulations (41) and
new drugs, such as pregabalin, which has a novel mechanism of action and negligible propensity for drug interactions (25), there is reason to expect options to increase
and outcomes to improve for both monotherapy and combination therapy in patients with epilepsy.
REFERENCES
1. Sander JW. The epidemiology of epilepsy revisited. Curr Opin Neurol 2003;16:16570.
2. Stefan H, Halasz P, Gil-Nagel A, et al. Recent advances in the diagnosis and treatment of epilepsy. Eur J Neurol 2001;8:51939.
3. Loiseau P. Seizure precipitants. In: Engel Jr J, Pedly TA, eds.
Epilepsy: a comprehensive textbook. Philadelphia: Lippincott Raven
Publishers, 1997:936.
4. Kwan P, Brodie MJ. Refractory epilepsy: a progressive, intractable
but preventable condition? Seizure 2002;11:7784.
5. Perucca E, Beghi E, Dulac O, et al. Assessing risk to benefit ratio
in antiepileptic drug therapy. Epilepsy Res 2000;41:10739.
6. Kwan P, Brodie MJ. Early identification of refractory epilepsy. New
Engl J Med 2000:342:3149.
7. Brodie MJ, Kwan P. The star systems: overview and use in determining antiepileptic drug choice. CNS Drugs 2001;15:112.
8. Leppik IE. Monotherapy and polypharmacy. Neurology 2000;
55(suppl 3):S259.
9. Beghi E, Perucca E. The management of epilepsy in the 1990s:
acquisitions, uncertainties and priorities for future research. Drugs
1995;49:68094.
10. Perucca E. Clinical pharmacology and therapeutic use of the new
antiepileptic drugs. Fundamental Clin Pharmacol 2001;15:40517.
11. Brodie MJ, French JA. Management of epilepsy in adolescents and
adults. Lancet 2000;356:3239.
12. Kwan P, Brodie MJ. Epilepsy after the first drug fails: substitution
or add-on? Seizure 2000;9:4648.
13. Dreifuss FE. Classification of epileptic seizures. In: Engel Jr J, Pedly
Epilepsia, Vol. 45, Suppl. 6, 2004

21.
22.
23.
24.
25.
26.
27.

28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.

TA, eds. Epilepsy: a comprehensive textbook. Philadelphia: Lippincott Raven Publishers, 1997:51724.
Sander JWAS, Sillanpaa M. Natural history and prognosis. In: Engel
Jr J, Pedly TA, eds. Epilepsy: a comprehensive textbook. Philadelphia: Lippincott Raven Publishers, 1997:6985.
Foldvary N, Wyllie E. Epilepsy. In: Goetz CG, Pappert EJ, eds.
Textbook of clinical neurology. Philadelphia: WB Saunders Co.,
1999:105988.
MacDonald BK, Johnson AL, Goodridge DM, et al. Factors predicting prognosis of epilepsy after presentation with seizures. Ann
Neurol 2000:48:83341.
Perucca E, Dulac O, Shorvon S, et al. Harnessing the clinical potential of antiepileptic drug therapy: dosage optimisation. CNS Drugs
2001;15:60921.
Richens A. Rational polypharmacy. Seizure 1995;4:2114.
Stephen LJ, Brodie MJ. Seizure freedom with more than one
antiepileptic drug. Seizure 2002;11:34951.
Gilliam F. Optimizing health outcomes in active epilepsy. Neurology
2002;58(8 suppl 5):S920.
Besag FMC, Berry DJ, Pool F, et al. Carbamazepine toxicity
with lamotrigine: pharmacokinetic or pharmacodynamic interaction? Epilepsia 1998;39:1837.
Pisani F, Oteri G, Russo MF, et al. The efficacy of valproate-lamotrigine comedication in refractory complex partial seizures: evidence
for a pharmacodynamic interaction. Epilepsia 1999;40:11416.
Patsalos PN, Froscher W, Pisani F, et al. The importance of drug
interactions in epilepsy therapy. Epilepsia 2002;43:36585.
Deckers CL. Place of polytherapy in the early treatment of epilepsy.
CNS Drugs 2002;16:15563.
Ben-Menachem E. Pregabalin pharmacology and its relevance to
clinical practice. Epilepsia 2004;45(suppl 6):138.
Krauss GL, Brandt J, Campbell M, et al. Antiepileptic and oral
contraceptive interactions: a national survey of neurologists and obstetricians. Neurology 1996;46:15439.
Shorvon SD, Tallis RC, Wallace HK. Antiepileptic drugs: coprescription of proconvulsant drugs and oral contraceptives: a national
study of antiepileptic drug prescribing practice. J Neurol Neurosurg
Psychiatry 2002;72:1145.
Johannessen SI, Henricksen O. Maintenance therapy. In: Engel Jr
J, Pedly TA, eds. Epilepsy: a comprehensive textbook. Philadelphia:
Lippincott Raven Publishers, 1997:124753.
Bazil CW. Comprehensive care of the epilepsy patientcontrol,
comorbidity, and cost. Epilepsia 2004;45(suppl 6):312.
Thapar AK, Stott NC, Richens A, et al. Attitudes of GPs to the care
of people with epilepsy. Fam Pract 1998;15:43742.
Chang S-I, McAuley JW. Pharmacotherapeutic issues for women of
childbearing age with epilepsy. Ann Pharmacother 1998;32:794
801.
Fisher RS, Vickrey BG, Gibson, P, et al. The impact of epilepsy from
the patients perspective I. Descriptions and subjective perceptions.
Epilepsy Res 2000;41:3951.
Jacoby A. Age-related considerations. In: Engel Jr J, Pedly TA,
eds. Epilepsy: a comprehensive textbook. Philadelphia: Lippincott
Raven Publishers, 1997:11219.
Austin JK, Smith MS, Risinger MW, McNelis AM. Childhood
epilepsy and asthma: comparison of quality of life. Epilepsia
1994;35:60815.
Hauser WA. Seizure disorders: the changes with age. Epilepsia
1992;33(suppl 4):S614.
Wallace HK, Solomon JK. Quality of epilepsy treatment and services: the views of women with epilepsy. Seizure 1999;8:817.
Fisher RS, Vickrey BG, Gibson P, et al. The impact of epilepsy
from the patients perspective II: views about therapy and health
care. Epilepsy Res 2000;41:5361.
Cramer JA, Glassman M, Rienzi V. The relationship between poor
medication compliance and seizures. Epilepsy Behav 2002;3:338
42.
Williams J, Myson V, Steward S, et al. Self-discontinuation of
antiepileptic medication in pregnancy: detection by hair analysis.
Epilepsia 2002;43:82431.
Commission on Classification and Terminology of the International
League Against Epilepsy. Epilepsia 1989;30:38999.
Nicolson A, Leach JP. Future prospects for the drug treatment of
epilepsy. CNS Drugs 2001;15:95568.