PROCEEDINGS
TREATMENT OF REFRACTORY EPILEPSY*
—
Barbara Olson, MD†
ABSTRACT
Despite advances in pharmacotherapy,
refractory epilepsy continues to be a significant
problem for many pediatric patients. No uniform
definition of intractable epilepsy exists. When
true pharmacoresistance is present, the ketogenic
diet, vagus nerve stimulation, and seizure
surgery are alternative therapeutic options.
Clinical considerations of various treatment
modalities in refractory childhood epilepsy are
A
discussed in this article.
(Adv Stud Med. 2005;5(5B):S470-S473)
pproximately 80 000 new cases of
epilepsy are diagnosed each year in chil-
dren. Fortunately, many of these young
patients achieve spontaneous remission;
but for those patients who do not, the
consequences of intractability can be quite severe,
including status epilepticus, cognitive decline, lowered
socioeconomic potential, and profound stress on the
entire family. Children and their families may encounter
significant problems in academic and social achieve-
ment and personal and family lifestyle plus financial
pressures caused by prolonged medical treatment.
Children with intractable seizures may also sustain sig-
nificant injuries caused by falls and other accidents.
*Based on a presentation given by Dr Olson at the
2004 Annual Meeting of the Child Neurology Society.
†Assistant Clinical Professor of Neurology and
Pediatrics, Vanderbilt University, Nashville, Tennessee.
Address correspondence to: Barbara Olson, MD, Pediatric
Associates, 2400 Patterson Street, Suite 216, Nashville, TN
37203. E-mail: bjolson@pedsneurology.com.
S470
The primary goal of epilepsy therapy is the absence
of seizures without treatment adverse effects. Some
antiepileptic drugs (AEDs), such as phenobarbital,
offer excellent antiseizure efficacy but have intolerable
adverse effects. Physicians caring for developing chil-
dren must be acutely aware of the safety implications
of these agents with regard to brain development, bone
health, and other issues—short term and long term.
AEDs can interact significantly with other drugs, as in
the case of their well-known potential for interaction
with birth control pills in our adolescent patients.
WHAT IS REFRACTORY EPILEPSY?
Existing medical literature shows that 10% to 30%
of newly treated patients with epilepsy do not respond
to medication, despite adequate treatment with appro-
priate AEDs. This article describes characteristics of
adequate antiseizure therapy, explores common rea-
sons for failure of AED treatment, and offers selected
caveats regarding monotherapy and combination ther-
apy with antiepileptic agents. Nonmedical treatment
options and common causes of drug resistance in
patients with epilepsy are also discussed.
A recent study involving 525 children and adults
demonstrated seizure resolution in approximately 50%
of all patients following treatment with a single AED.1
These patients’ chance of attaining remission dropped
precipitously as further antiseizure agents were used
unsuccessfully, as only 4% of the patients became
seizure-free after 3 AEDs were used, alone or in com-
bination. Overall, 33% of all patients in this study
eventually were found to have intractable epilepsy.
In a recent prospective study, 10% of 613 children
had intractable epilepsy, defined as the failure of 2 or
more AEDs over an 18-month period.2 An earlier lon-
gitudinal study involving 417 children with partial
and generalized tonic-clonic seizures indicated remis-
Vol. 5 (5B) n May 2005
 PROCEEDINGS
sion in 80% of patients who were able to remain on a
single AED for 1 year.3 In this group, 17% of the chil-
dren (typically those patients with underlying neuro-
logic deficits) required treatment with additional
AEDs. Overall, 25% of the children in this study who
did not respond to one AED had refractory epilepsy.
A recent international review involving 7 indepen-
dent studies of childhood epilepsy found that long-
term remission occurs in approximately 60% of
children overall.4 Factors favoring remission include
the absence of significant neurologic and intellectual
deficits, age younger than 12 years at seizure onset,
and the presence of infrequent, easily controlled
seizures. The presence of all these factors was associat-
ed with a remission rate of 80%; the remission rate fell
to 20% when these factors were absent.
EPILEPSY TREATMENTS
ANTIEPILEPTIC DRUGS
Since 1993, 9 new AEDs have been introduced in
the United States. As compared with the older drugs,
some of these drugs involve fewer adverse effects and
are easier to use, as blood levels are not determined as
often. Mechanisms of AED action are summarized in
Table 1.5
Carbamazepine and phenytoin and the newer
agents lamotrigine and oxcarbazepine affect the volt-
age-gated sodium channels and limit repetitive firing
of irritable neurons, whereas the calcium channel
blockers have their primary effect by raising the
threshold of the calcium channels. The inhibitory neu-
rotransmitter class includes the benzodiazepines and
phenobarbital, in addition to some of the newer
AEDs. The classification of AEDs with regard to
mechanism of action simplifies agent selection, as an
ineffective agent can be replaced by an agent with a
differing mechanism; agents of different classes can be
combined for increased therapeutic effect.
COMMON CAUSES OF ANTIEPILEPTIC DRUG FAILURE
Antiepileptic drug therapy may be ineffective for
many reasons. Correct diagnosis may be difficult at
times because physicians must often rely on second-
party and third-party observation of seizure events.
The diagnostic picture may be complicated by pseu-
doseizures and other factors and by comorbidities,
such as autism, spastic cerebral palsy, and repetitive tic
Advanced Studies in Medicine n S471
movements. Therefore, a video electroencephalogram
(EEG) may be used in many cases to identify exactly
what is happening with the patient.
Patients may be treated with a correct AED but are
given an inadequate dose or a too short a trial of the
medication, as patients with difficult-to-control
seizures often require AED doses near the maximum
limit of tolerability. Some newer agents are used suc-
cessfully at much higher than recommended doses but
obviously with careful attention paid to adverse-effect
outcomes. Physicians must understand that it is some-
times more effective to push a single AED to maxi-
mum tolerated levels than to add a second medication
in patients who have experienced 1 seizure on an ini-
tial AED. Failure to use the recognized drug of choice
for specific seizure syndromes can, of course, affect
treatment effectiveness.
Promoting treatment compliance is an important
part of treating childhood epilepsy and often comprises
a major issue in treatment effectiveness. If parents or
caregivers are not compliant with therapy, the causes of
this noncompliance must be carefully, and often deli-
cately, explored. AED noncompliance may be because of
family members’ fears about a drug’s long-term adverse
effects or the high cost of these agents, which may not be
covered by third-party payers. Furthermore, denial may
play a part when caregivers express concerns about the
effectiveness of or the need for specific AEDs. Pill boxes
and seizure calendars can help families improve treat-
ment compliance, and sometimes drug levels must be
Table 1. Mechanisms of Action of Antiepileptic
Drugs
Na+ Channels Ca2+ Channels
Carbamazepine Ethosuximide
Lamotrigine Topiramate
Oxcarbazepine Zonisamide
Phenytoin
Inhibitory Transmission
Benzodiazepines Sodium valproate
Felbamate Tiagabine
Gabapentin Topiramate
Phenobarbital
Ca2+ = calcium; Na+ = sodium.
Data from Kwan et al.5
 PROCEEDINGS
monitored and office-visit frequency increased when
more physician supervision is necessary.
Radiologic data should be re-evaluated carefully to
rule out structural causes for AED ineffectiveness.
Very high-quality, high-resolution magnetic resonance
imaging studies are necessary, and any questionable
studies should be repeated. Also, healthcare providers
must be aware that orthodontic braces can cause arti-
factual findings in radiologic studies done in children,
thus these braces must sometimes be temporarily
removed and the original studies repeated to confirm
that nothing has been missed or misinterpreted.
Physicians are aware that certain seizure syndromes
may be inherited within families, and recent excellent
work in pharmacogenomics has made it clear that drug
resistance can also be associated with certain geno-
types. Future work in this area may lead to genetic
tests that are of practical use in identifying patients
who are genetically resistant to certain AEDs.
ANTIEPILEPTIC DRUG TREATMENT STRATEGIES
In selecting single and multiple treatment agents,
physicians must carefully consider the potential effect of
comorbidities, such as mood instability, headaches, and
other factors, that commonly accompany childhood
epilepsy. AED monotherapy offers clear advantages,
especially in children, with regard to fewer adverse effects
and drug-to-drug interactions, better patient compli-
ance, and lower cost. However, when an initial AED
proves ineffective, an appropriate second agent can be
added and the initial drug withdrawn after an evaluation
has been made of the effect of the 2 drugs together. It is
essential during this process to carefully obtain feedback
from the family about the drug’s effect on the child’s
seizures and to note any adverse effects or problems
when each agent is administered alone and when both
agents are administered together. Table 2 summarizes
AED combinations that may be helpful in refractory
forms of epilepsy.6-9
Studies have shown that ethosuximide and val-
proate are effective in treating absence seizures.6
Lamotrigine must be used carefully in combination
with valproate in treatment of combination partial and
generalized seizures because of the increases in drug
half-life, although this effect may at times be advanta-
geous.7,8 Lamotrigine plus topiramate is effective for
combination partial and generalized seizures and in
the treatment of partial seizures alone.9
S472
PREDICTING DRUG RESISTANCE
Approximately 10% to 30% of children with
epilepsy exhibit some degree of resistance to AEDs,
defined as the failure of 2 or more agents. In one study,
AED resistance was associated with high initial seizure
frequency, syndromic groupings, and significant focal
slowing on EEG.2 Another investigation identified
early myoclonic encephalopathy, early infantile epilep-
tic encephalopathy, severe myoclonic epilepsy in infan-
cy, and the presence of Lennox-Gastaut syndrome or
infantile spasms as indicators associated with increased
rates of AED resistance.10 High initial seizure frequen-
cy is associated with drug resistance, as are mixed
seizure types, early seizure breakthrough, abnormal
neurologic status, neonatal seizures, and the presence
of status epilepticus and tumors or other underlying
structural abnormalities.11
NONMEDICATION OPTIONS
KETOGENIC DIET
The ketogenic diet, which was developed more
than 80 years ago, controls seizure activity by a mech-
anism that has not yet been identified. The dietary
changes involved are complicated and require exten-
sive family commitment, but they may be extremely
effective in seizure reduction.12 The diet includes
80% to 90% of calories from fat, protein appropriate
for growth, and extreme carbohydrate restriction; the
diet typically is more successful with younger chil-
dren in whom diet is easily controlled by parents.
Potential adverse effects of the ketogenic diet include
lethargy, weight loss, nausea and vomiting, constipa-
tion, and diarrhea.
Table 2. Combination Therapy in Childhood
Epilepsy
Combination Seizure Type
Ethosuximide + sodium valproate Absence6
Lamotrigine + sodium valproate Combine partial/generalized7,8
Lamotrigine + topiramate Combine partial/generalized;
generalized9
Data from Rowan et al6; Pisani et al7; Brodie and Yuen8; and Stephen et al.9
Vol. 5 (5B) n May 2005
 PROCEEDINGS
Furthermore, the diet’s use necessitates the frequent
monitoring of complete blood count levels, electrolyte
values, and liver and renal status, as additional infre-
quent adverse effects can include hyperlipidemia, hypo-
glycemia, hypocalcemia, electrolyte imbalances, and
metabolic acidosis, in addition to cardiac and renal
abnormalities. Families interested in this option should
be advised to read Dr John Freeman’s book on epilepsy
diet treatment to gain a clear idea of the commitment,
responsibilities, and potential adverse effects involved in
this approach.13 Less restrictive ketogenic diets, such as
the modified Atkins diet,14 have also demonstrated
promise regarding seizure reduction and are the subject
of ongoing research.
EPILEPSY SURGERY
Surgery continues to be used as a treatment for
refractory pediatric epilepsy. Appropriate surgical
candidates have clearly demonstrated refractory
epilepsy (a localized, potentially removable area of
epileptogenic focus) and must be at low risk for new
neurologic deficits caused by the surgery. Before pur-
suing any surgical approach to treatment, families
must understand that surgical intervention is a long-
term process requiring a consistent clinical history,
positive radiologic and neurophysical evidence in
support of surgical intervention, and favorable neu-
ropsychologic findings. After this careful screening,
success rates of surgical intervention can be signifi-
cant, with remission rates of 60% to 80% after ante-
rior temporal lobectomy, and as high as 70% after
lesional extratemporal resection, hemispherectomy,
and multilobar resection.15
CONCLUSIONS
Physicians who care for children with refractory
epilepsy, in addition to their families, cannot afford to
be pessimistic. As the pioneering neurologist William
Lennox, MD, observed in 1928, “One who is con-
fronted with the task of controlling seizures in a per-
son with epilepsy grasps at any straw.” Pediatric
neurologists face difficult tasks every day in the treat-
ment of refractory epilepsy and must make the best use
Advanced Studies in Medicine n S473
of all available “straws,” including drug and diet ther-
apy, vagus nerve stimulation, and surgical interven-
tion, to help children with refractory epilepsy
approach a better life.
REFERENCES
1. Kwan P, Brodie MJ. Early identification of refractory epilep-
sy. N Engl J Med. 2000;342:314-319.
2. Berg AT, Shinnar S, Levy SR, et al. Early development of
intractable epilepsy in children: a prospective study [pub-
lished correction appears in Neurology. 2001;57:939].
Neurology. 2001;56:1445-1452.
3. Camfield C, Camfield P, Gordon K, et al. Outcome of
childhood epilepsy: a population-based study with a simple
predictive scoring system for those treated with medication.
J Pediatr. 1993;122:861-868.
4. Camfield P, Camfield C. Childhood epilepsy: what is the
evidence for what we think and what we do? J Child
Neurol. 2003;18:272-287.
5. Kwan P, Sills GJ, Brodie MJ. The mechanisms of action of
commonly used antiepileptic drugs. Pharmacol Ther.
2001;90:21-34.
6. Rowan AJ, Meijer JW, de Beer-Pawlikowski N, et al.
Valproate-ethosuximide combination therapy for refractory
absence seizures. Arch Neurol. 1983;40:797-802.
7. Pisani F, Oteri G, Russo, MF, et al. The efficacy of val-
proate-lamotrigine comedication in refractory complex par-
tial seizures: evidence for a pharmacodynamic interaction.
Epilepsia. 1999;40:1141-1146.
8. Brodie MJ, Yuen AW. Lamotrigine substitution study: evi-
dence for synergism with sodium valproate? 105 Study
Group. Epilepsy Res. 1997;26:423-432.
9. Stephen LJ, Sills GJ, Brodie MJ. Lamotrigine and topiramate
may be a useful combination. Lancet. 1998;351:958-959.
10. Dunn DW, Buelow JM, Austin JK, et al. Development of syn-
drome severity scores for pediatric epilepsy. Epilepsia.
2004;45:661-666.
11. Kwong KL, Sung WY, Wong SN, So KT. Early predictors of
medical intractability in childhood epilepsy. Pediatr Neurol.
2003;29:46-52.
12. Vining EP, Freeman JM, Ballaban-Gil K, et al. A multicenter
study of the efficacy of the ketogenic diet. Arch Neurol.
1998;55:1433-1437.
13. Freeman JM, Kelly MT, Freeman JB. The Epilepsy Diet
Treatment: An Introduction to the Ketogenic Diet. 3rd ed.
New York, NY: Demos Vermande; 2000.
14. Kossoff EH. More fat and fewer seizures: dietary therapies
for epilepsy. Lancet Neurol. 2004;3:415-420.
15. Wyllie E, Comair YG, Kotagal P, et al. Seizure outcome
after epilepsy surgery in children and adolescents. Ann
Neurol. 1998;44:740-748.