Management of Epilepsy Following Initial Diagnosis

Management of Epilepsy Following Initial Diagnosis:
Current Best Practices

James Thomas Houston, MD

University of Alabama at Birmingham, Birmingham, Alabama

Management of epilepsy involves careful and prompt diagnosis, selection of appropriate initial medical therapy, close patient monitoring, and, if needed, surgical intervention. Speakers at this symposium covered failure of drug therapy, best use of drug combinations, and novel nonpharmacologic treatments that may be tried in patients with refractory epilepsy. In addition, the concept of disease-modifying therapies for epilepsy was discussed.

James Thomas Houston, MDControl of seizures represents a balance between medication management and clinical decisions designed to circumvent drug failure. During a symposium titled "Epilepsy Therapy: What You Need to Know to Get Your Patients Into the 65% Group," speakers at the 2015 annual meeting of the American Epilepsy Society described methods to manage epilepsy from the patient's diagnosis through initial selection of antiepileptic drugs (AEDs), use of nonpharmacologic therapies, and consideration of surgery to limit seizure activity. The panel also addressed specific problems in controlling seizures and pointed to new avenues of research that certainly will be useful in the future.

The symposium was chaired by Cynthia Harden, MD, Director of Epilepsy Services at the Mount Sinai Health System in New York City, and Jerry Shih, MD, Director of the Comprehensive Epilepsy Program at the Mayo Clinic in Jacksonville, Florida, and Associate Professor of Neurology at the Mayo College of Medicine in Jacksonville.

Based on a presentation by Emilio Perucca, MD, PhD, Professor of Clinical Pharmacology, University of Pavia, and Director, Clinical Trial Center, C. Mondino National Neurological Institute, IRCCS, Pavia, Italy.

Confirmation of epilepsy is the most important step to take before prescribing any AED. An accurate diagnosis of epilepsy can be achieved using any of the following criteria1:

  • One unprovoked seizure and a high probability of more seizures occurring over the following 10 years, as evidenced by the finding of structural abnormalities on magnetic resonance imaging;
  • Two unprovoked seizures occurring 24 hours or more apart; or
  • Diagnosis of an epilepsy syndrome.

Timing of treatment initiation, while essential to prevent seizure recurrence in high-risk patients, does not reduce the long-term probability of seizure freedom. Results of the FIRST trial showed that delaying treatment until occurrence of a second seizure did not lower the probability of seizure freedom at 1–4 years.2 Given these data, monitoring of low-risk patients for development of epilepsy without starting treatment may prevent possible adverse reactions from unnecessary therapy.

Selection of an Appropriate AED
Appropriate treatment of newly diagnosed epilepsy is a key component of long-term seizure outcomes. Approximately 50% of patients achieve seizure freedom using the first AED prescribed (Figure 1).3 Individualizing both drug choice and dosing schemes helps provide maximal patient benefit and limits development of disabling side effects. Selection of an appropriate initial AED involves consideration of its spectrum of efficacy, mechanism of action, side-effect profile, ease of administration, cost, drug interactions, and potential impact on existing comorbidities.

Figure 1

FIGURE 1 Response to antiepileptic drug medication in 470 previously untreated patients with epilepsy. Adapted, with permission, from Kwan and Brodie.3

There is considerable overlap in the efficacy spectrum of many currently approved AEDs (Table 1).4 A significant number of these medications have specific indications and contraindications that must be considered when physicians decide upon a therapy. Generally, most antiepileptics designed and approved for focal epilepsies provide additional protection against generalized tonic-clonic seizures, but their use may exacerbate absence and myoclonic seizures.

Table 1

Among the AEDs approved for initial treatment of focal epilepsies, carbamazepine, phenytoin, oxcarbazepine, lamotrigine, levetiracetam, and zonisamide have similar efficacy profiles. Lamotrigine, while effective in treating multiple types of epilepsy, may worsen seizures in infants with severe myoclonic epilepsy.

For treating new-onset focal epilepsy, newer generation AEDs generally have not been shown to be more effective than older AEDs, such as carbamazepine, valproate, and phenytoin. Gabapentin and vigabatrin have been shown to be the least effective of the drugs evaluated in this situation.5,6

Phenobarbital and primidone are effective against most seizure types, but they are ineffective for treating absence seizures and are less effective than carbamazepine, phenytoin, or valproic acid for treating local epilepsies.7

Use of AEDs in Specific Populations of Patients With Epilepsy
Multiple studies have found valproate to be the most effective AED against idiopathic generalized epilepsy, juvenile myoclonic epilepsy, and unclassified epilepsy; this finding also held true during direct comparisons of valproate with lamotrigine and topiramate. Whereas valproate and ethosuximide therapy have similar success rates in treating childhood absence seizures, ethosuximide remains the preferred treatment. Valproate has more potential for causing adverse effects, and ethosuximide has a greater potential to modify the properties of epilepsy. Despite numerous controlled trials examining the initial treatment of new-onset epilepsy, class 1 evidence—evidence based on the outcomes of randomized controlled clinical trials—exists only for focal-onset epilepsies and childhood absence epilepsy.8

To ensure maximal benefit with the lowest possible risk, other factors must be considered before initial AED therapy is selected. Genotype, previous adverse drug reactions, age, gender, comorbidities, and concomitantly administered medications all are potential risk factors for adverse reactions. For example, carbamazepine should be avoided in patients considered to be carriers for the specific alleles human leukocyte antigen (HLA)-B 15:02 and HLA-A 31:01, whereas the aromatic AEDs should be used cautiously in patients with a history of significant drug hypersensitivities. AEDs with a higher occurrence of cognitive side effects should be avoided or used sparingly in patients of advanced age, and school-aged children and adolescents should be followed closely.

Use of AEDs During Pregnancy
When considering antiepileptic therapy, women of childbearing age represent a unique, challenging group of patients. The estimated rate of major malformations in the general population of North America ranges from 1% to 3%. Data from Hernández-Díaz et al9 revealed that among pregnant women taking AEDs, valproate was associated with the highest rate of major malformations (9.3%), followed by phenobarbital (5.5%) and topiramate (4.2%). Taking clonazepam, carbamazepine, or phenytoin during pregnancy resulted in a similar risk (3%); treatment with levetiracetam, lamotrigine, oxcarbazepine, gabapentin, and zonisamide was associated with the least risk, ranging from 2.4% to as little as 0% (Figure 2).9

Figure 2

FIGURE 2 Major malformation rates in monotherapy exposures: North American Registry Data. Adapted, with permission, from Hernández-Díaz.9

Additionally, the risk of offspring malformations in pregnant women taking carbamazepine, lamotrigine, phenobarbital, or valproate rises significantly with increasing doses of these medications.10 Selection of an appropriate AED in this important patient population therefore demands consideration of malformation risk versus seizure control, determination of minimal effective dosing, and patient access to serum drug-level monitoring.

Therapeutic Titration
After an appropriate AED has been selected, determining the initial target dose and titration schedule are essential to maximize early seizure control and minimize the risk of dose-related adverse effects. Kwan and Brodie11 showed that moderate doses of carbamazepine (400–600 mg/d), valproate (600–1,000 mg/d), and lamotrigine (125–200 mg/d) are the most effective in achieving seizure freedom. Low-dose therapy provided minimal efficacy, whereas high-dose therapy resulted in decreasing efficacy with successive up-titrations. In a study comparing levetiracetam and carbamazepine monotherapy, Brodie and others12 showed that of those patients who achieved a 1-year remission, 89% of the patients taking carbamazepine achieved it on only 400 mg/d. Similarly, of those patients achieving 1-year seizure freedom on levetiracetam, 86% did it on only 1,000 mg/d. Although a small percentage of patients will achieve improved seizure control on doses above the recommended range, low-dose titration should be the goal when starting AED therapy.

There are numerous factors involved in the initiation of antiepileptic therapy. Proper diagnosis, an individualized treatment plan, and achieving a risk/benefit balance are essential for developing an appropriate and effective treatment plan for any newly diagnosed epilepsy patient.

Based on a presentation by Patrick Kwan, MD, PhD, Professor of Neurology, University of Melbourne, and Head of Epilepsy, Royal Melbourne Hospital, Melbourne, Australia.

The many proposed definitions of drug-resistant epilepsy (DRE) can be summarized as failure of an adequate trial of two or more AEDS that were tolerated and appropriately chosen and used.13 Approximately 36% of newly diagnosed epilepsy patients become pharmacoresistant.11 Of patients who meet these criteria, < 20% can be expected to become seizure-free with additional medications, and a significant number relapse after 1 year of seizure freedom.

Before considering further treatment options, true DRE must be differentiated from so-called pseudoresistance, which may result from misdiagnosis, an inappropriate drug choice, inadequate dosing, a multitude of lifestyle issues, and other factors.

Subtherapeutic dosing is one of the most significant and easily preventable causes of pseudoresistance. Investigations into initial monotherapy with carbamazepine, lamotrigine, and valproate have revealed that the majority of patients (~ 80%) who achieved seizure freedom did so on the defined daily dose (DDD) or less; other patients became seizure-free on increasingly larger doses.11 Similar results have been shown for levetiracetam, especially in patients showing a partial response to doses at or below the DDD.

Unfortunately, in clinical practice, a second agent often is added to an uncontrolled patient's regimen before a dose escalation above the DDD is attempted. Subsequent failure of this second agent may lead to an incorrect diagnosis of DRE and unnecessary evaluation for additional therapies. Unless contraindicated due to intolerability, therefore, dose escalation above the DDD should be considered in all newly diagnosed patients, especially those showing partial response to lower doses of the initial AED.

Poor patient compliance and drug/alcohol abuse are among the most important modifiable lifestyle factors contributing to pseudoresistance. Patients who are highly suspected of noncompliance or substance abuse may need to be evaluated for untreated psychiatric illness, closer monitoring of AED blood levels, and determination of adequate financial support. Failure to recognize these contributing factors may lead to inappropriate treatment decisions and poor long-term seizure outcomes.

If the failure of using two AEDs is confirmed with any degree of confidence, diagnosis of true DRE also requires a complete diagnostic review and reevaluation of seizure type/localization before adding another drug is considered. Misdiagnosis of psychogenic nonepileptic seizures, failure to recognize cardiac disorders, or false classification of underlying seizure types can lead to poor outcomes. For example, incorrect diagnosis of focal epilepsy in a patient with idiopathic generalized epilepsy may result in inappropriate AED selection and greater seizure frequency. Referral to a tertiary epilepsy center may be needed for proper secondary evaluation.

Once a confident diagnosis of DRE has been established and all potentially modifiable risk factors for pseudoresistance have been excluded, two options are available to increase the likelihood of seizure freedom: "rational" polypharmacy and nonpharmacologic therapy.

Based on a presentation by Josiane LaJoie, MD, Associate Professor of Neurology and Pediatrics, NYU Langone Medical Center, New York, New York.

The potential benefit of polytherapy in treating epilepsy has been known for over 150 years. The use of bromide with other medications as having greater efficacy than monotherapy was first described by Gowers in 1881.14 Early studies of older-generation AEDs showed that combination therapy was superior in efficacy and toxicity than was monotherapy, with combinations of drugs having different mechanisms of actions (eg, phenobarbital/phenytoin) being more effective than polytherapy using AEDs with similar mechanisms of action (eg, phenytoin/carbamazepine).

These results were so intriguing at the time that pharmaceutical companies began developing medications that combined two AEDs in one pill. However, use of these combination AEDs slowly decreased as newer-generation drugs were introduced during the 1980s, and novel monotherapies were favored over the use of older AED polytherapies. Current evidence favors the use of "rational polypharmacy," including combining both newer- and older-generation drugs.

When initiating polytherapy for patients with true DRE or those who have experienced failure after taking adequate doses of multiple monotherapies, a combination of AEDs with supra-additive effects and limited combined adverse effects is most effective. In addition, patient quality of life (QOL) and underlying condition should show improvement overall to justify polytherapy.

Polytherapies based on mechanisms of action may enhance antiseizure effects. Certain combinations appear to offer increased effectiveness, whereas others may have limited synergistic effects. For example, sodium-channel blockers used with γ-aminobutyric acid (GABA) inhibitors may provide increased benefit, whereas the combined use of two sodium-channel agents may show less promise. Using two GABA-mimetic drugs or an α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) antagonist with an N-methyl-D-aspartate antagonist also may provide greater benefit, with the latter potentially having intolerability issues. Some results from laboratory studies have suggested that the most successful drug combinations involve the use of a single-mechanism drug with an agent possessing multiple mechanisms of action.

Adding lamotrigine to valproate in patients with both partial and idiopathic generalized epilepsy (IGE) produced a significantly higher response rate (64%) than was found when lamotrigine was added to either phenobarbital or carbamazepine (41%) or phenytoin (38%).15 Curiously, when lamotrigine was substituted as monotherapy for responders, IGE patients had a higher responder rate than did patients with partial-onset epilepsy. Lamotrigine monotherapy provided more benefit than did some combinations, and long-term follow-up studies revealed that coadministration of lamotrigine and valproate was superior in efficacy to use of either drug alone.

In a retrospective review of both partial and primary generalized epilepsy, Stephen and colleagues16 found that about 80% of patients were controlled on two AEDs, with valproate and lamotrigine being the most effective. A significant percentage of the remaining patients were controlled on three AEDs, with the most successful combinations being valproate and lamotrigine plus either topiramate or levetiracetam.

AED Combinations for Specific Pediatric Seizure Disorders
Use of many drug combinations has shown specific efficacy depending upon the underlying seizure disorder.

Severe myoclonic epilepsy of infancy (SMEI). In SMEI, a malignant epilepsy syndrome first noted during the first year of life, valproic acid plus clobazam has been effective, whereas use of vigabatrin, lamotrigine, and carbamazepine could potentially worsen seizures. Stiripentol is a direct allosteric modulator of the GABA-A receptor that has not been approved by the US Food and Drug Administration (FDA). Chiron and colleagues17 showed that using this drug with valproate and clobazam provided a significant synergistic effect and increased responder rate in SMEI patients. Adding topiramate to either valproate or valproate plus clobazam and stiripentol may reduce the frequency of generalized tonic-clonic seizures and the risk of status epilepticus. These findings suggest that stiripentol may provide a synergistic effect with clobazam in treating individuals with uncontrolled SMEI, and adding topiramate to various polytherapies may reduce the total frequency of generalized tonic-clonic seizures. Of note, the use of stiripentol with topiramate should be instituted cautiously, since both inhibit the activity of cytochrome P450.

Lennox-Gastaut syndrome (LGS). In treating LGS, valproate typically is used as a first-line therapy, but several other agents may provide adjunctive benefit. The addition of felbamate may be particularly effective in controlling atonic seizures, whereas the addition of topiramate and lamotrigine may have an overall synergistic effect. In patients receiving valproate, adding clobazam provided a significant decrease in drop seizures when compared with placebo, and more benefit was found at increasingly higher doses. Brodie et al18 also suggested that adding rufinamide to valproate, lamotrigine, gabapentin, phenytoin, or phenobarbital may provide additional benefit for reducing the frequency of partial seizures, but adding rufinamide to regimens containing carbamazepine was no more beneficial than the addition of placebo. Further study is needed, however, to confirm these findings.

A review of all controlled trials of AEDs in patients with LGS may be summarized as follows: lamotrigine and felbamate (and topiramate, to a lesser extent) may be beneficial adjunctive therapies, especially when treating atonic seizures. Use of lamotrigine and felbamate has been more effective in decreasing generalized tonic-clonic seizures than has topiramate administration, and the addition of rufinamide to AED regimens that do not contain carbamazepine may provide additional benefit in reducing the frequency of all seizure types. A ketogenic or other diet combined with vagal-nerve stimulation may have additional benefit in this patient population.

Other seizure types. Other potential combinations of AEDs may have increased efficacy in patients with childhood absence seizures, partial seizures, myoclonic epilepsies, and juvenile myoclonic epilepsy. Administration of valproate with ethosuximide may have a synergistic effect in treating childhood absence seizures. Addition of vigabatrin or valproate to carbamazepine therapy may provide a greater reduction in seizure frequency in patients with partial epilepsies. Also, administration of vigabatrin with tiagabine may have synergistic effects in treating partial seizures, and valproate plus levetiracetam is an effective dual therapy in reducing the frequency of myoclonic seizures. Lastly, combinations of lamotrigine with either topiramate or valproate are considered to be good choices for juvenile myoclonic epilepsy polytherapy.

In summary, many polypharmacy options are available to treat various underlying epilepsies, but certain adjunctive or combination therapies should be considered, depending upon the specific epilepsy disorder and the patient's current ongoing regimen.

Based on a presentation by Christopher T. Skidmore, MD, Assistant Professor of Neurology, Sidney Kimmel College of Medicine, Thomas Jefferson University, and member of the Jefferson Comprehensive Epilepsy Center, Philadelphia, Pennsylvania.

If an epileptic patient does not respond to the concomitant use of three AEDs, and adverse effects become an issue when other agents are added, nonpharmacologic treatment options should be considered. These options include dietary modification, responsive neurostimulation, vagal-nerve stimulation, and deep-brain stimulation.

Dietary Modification
Dietary modification provides a 50% reduction in seizure frequency in approximately half of patients trying this treatment option. Three possible dietary-modification regimens may reduce seizure frequency: the ketogenic diet, the modified Atkins diet, and the low glycemic-index diet. Formulations of dietary intake are specific for each regimen.19

The standard US diet consists of 50% carbohydrates, 15% protein, and 35% fat. The ketogenic diet requires fat consumption to be 90% of the patient's diet. The modified Atkins diet allows for 5%–10% carbohydrate consumption, with fat intake consisting of 60%–65% of the overall regimen. The low glycemic-index diet allows for 20%–30% carbohydrate consumption, with fat consumption consisting of 60%–70% of overall intake.

Specific differences between the modified Atkins and ketogenic diets should be considered when deciding between these two options (Table 2).19 The modified Atkins diet offers unrestricted daily caloric and fluid intake, whereas the ketogenic diet has caloric and fluid restrictions. The ketogenic diet typically requires hospital admission, a fasting period, intensive education, and involvement of a dietician; these parameters are not required by the modified Atkins regimen. In addition, food-sharing by family, the ability to eat some restaurant foods, and the use of some low-carbohydrate store-bought products are allowed under the modified Atkins diet, but they are not approved for the ketogenic diet. Therefore, overall QOL appears to be less affected with use of the modified Atkins diet than with the ketogenic diet and must be considered when an appropriate dietary option is being selected.

Table 2

The healthcare professional also must assess the patient's baseline cognitive status and intellectual capacity before choosing a specific diet regimen. Potential side effects (eg, constipation, elevated lipids, weight loss) also must be considered. Notably, both diets have shown similar efficacy in reducing seizure frequency.

Neurostimulation involves three potential modalities. Vagal-nerve stimulation received the first FDA approval in 1997 and is indicated for treatment of refractory focal epilepsies in patients > 12 years of age. Responsive neurostimulation was approved in 2013 specifically for refractory focal epilepsies in adult patient populations; it typically is instituted in patients with drug-resistant partial seizures who are not good surgical candidates and who have one or two identified seizure foci. Deep-brain stimulation focuses on the anterior nucleus of the thalamus; it was approved in Europe in 2010 and Canada in 2012 for treatment of refractory focal epilepsy in adults, but it has yet to be approved in the United States.

Vagal-nerve stimulation. This open-loop feedback system may be used with magnetic stimulation for on-demand applications. Potential side effects include cough and hoarseness, which typically occur as the output current is increased. Stimulation frequency and duration are adjusted independently for each patient, and changes in settings require a qualified practitioner's involvement.

Newer modifications may provide automated stimulation depending upon changes in heart rate that may accompany ictal activity. In this newer system, stimulation triggers are based on the percent change in heart rate (20%–70%), with baseline heart rate configured using a slow-moving average. Practitioners are required to set trigger thresholds before the autostimulation mode is used. Clinical evaluations of this proposed feature have shown that lowering the threshold for stimulation increases the sensitivity for recognizing ictal activity and also causes a proportional rise in false-positive findings.20

Essentially, setting the device to fire at 70% of the predetermined baseline heart rate is related to the lowest false-positive firing rate and the lowest sensitivity in detecting seizure onset. Setting the device threshold to fire at 20% of baseline heart rate is likely to detect the majority of seizures, but it also may incur numerous unwanted stimulations that may negatively affect patient compliance and QOL.

In summary, vagal-nerve autostimulation offers feasibility and safety comparable to those of the normal device, but its likelihood to improve clinical outcomes has not yet been determined.

Responsive neurostimulation. This closed-loop system involves placement of a four-contact cortical strip with a second cortical strip or, alternatively, a depth electrode that also has four contacts. These two electrodes are attached to a skull-implanted loop recorder that is involved in seizure pattern recognition and data compilation.

Morrell et al21 studied 191 patients with matched demographic data who were randomized to a sham- or active-treatment group. Both groups underwent electrode/strip placement. The sham-treatment group began responsive neurostimulation 5 months after surgery. Both groups had an initial 1-month period of improved seizure control without the device providing neurofeedback stimulation, which may suggest a localized microlesional effect after electrode placement (Figure 3).21

Figure 3

FIGURE 3 Mean number of disabling seizures by months postimplant among patients given neurofeedback stimulation or sham therapy. Adapted, with permission, from Morrell.21

Over the first 4 months after the device was turned on, the treatment group experienced a steady, significant reduction in seizure frequency when compared with the sham-treatment group, which showed a steady increase in seizures back to pretreatment baseline after the initial 1-month "honeymoon" period. When the sham-treatment group's stimulators were turned on at 4 months, seizure frequency improved, although less dramatically than seen in the active-treatment group over the same period. Long-term follow-up over several years has shown possible continued improvement in seizure frequency and responder rates over time, suggesting that responsive neurostimulation actually may modify epilepsy.

Responsive neurostimulation holds great promise as an additional treatment option for refractory patients. Potential side effects are similar to other deep-brain stimulation procedures (eg, for Parkinson's disease) and include implant-site infection and rare intracranial hemorrhage.

Deep-brain stimulation. Use of deep-brain stimulation has not been approved as an ancillary treatment for epilepsy in the United States. However, this novel therapy may prove to be a future therapeutic option.

The design of the Stimulation of the Anterior Nucleus of the Thalamus for Epilepsy (SANTE) trial22was similar to that of the initial responsive neurostimulation study, in that patients were matched for demographic information and then randomized to sham or control groups. Both then received deep brain-stimulation electrode implantation into the anterior thalamic nucleus. Both groups had an initial 1-month "honeymoon" period of reduced seizure frequency that was strikingly similar to the initial findings of the responsive neurostimulation studies.

Although microlesional theories seem intuitive for responsive neurostimulation, it is not known why there is an initial reduction in seizures independent of seizure localization or seizure type after thalamic electrode implantation. Thus, thalamic involvement in seizure networks and propagation must be studied further.

Results of short-term follow-up studies have been similar for responsive neurostimulation and deep-brain stimulation—patients in the sham-treatment group returned to baseline seizure frequency over a 4-month period when compared with the active-treatment group, which experienced a steady decrease in seizure frequency.22 Results of long-term follow-up studies have shown a significantly greater mean reduction of seizure frequency at 5 years than at 1 year, which holds true for responder rates as well. These results suggest that responsive neurostimulation and deep-brain stimulation are similarly effective at 1 year and show greater long-term benefit at 5 years and beyond. Potential complications of deep-brain stimulation are similar to those of responsive neurostimulation implantation, although a significant number of patients who receive deep-brain stimulation complain of paresthesias.

Disease Modification
Ultimately, long-term follow-up will determine whether either option is superior in terms of seizure reduction/response or a greater potential for disease-modifying properties. Disease modification represents a change in the expression or course of a disease or in comorbidities integral to the underlying disease. Disease modification is a treatment goal, but demonstration of actual disease modification in epilepsy remains challenging. Efficacy of AEDs can be a confounder, and many diverse endpoints are amenable to modification. Therefore, options beyond antiseizure treatment must be considered when developing strategies to achieve disease modification.

Long-term follow-up studies of the SANTE trial show continued steady improvement in mean seizure frequency at 5 years, which apparently was consistent across all patients receiving stimulation, regardless of underlying seizure type. This suggested, but failed to confirm, disease modification.22 A review of seizure outcomes after battery depletion in patients undergoing deep-brain stimulation showed baseline seizure frequency to be considerably lower than that seen preimplantation, although it was higher than that at baseline when the stimulation battery was operational (Table 3).23 This suggested possible disease modification, given the fewer seizures recorded following battery depletion as compared with preimplantation baselines.

Table 3

Long-term follow-up and monitoring will determine whether seizure activity in these patients continues to improve as battery life again expires. Similar studies should be considered in responsive neurostimulation patients to determine whether seizure activity improves after battery depletion when compared with baseline data. These findings would suggest both disease modification and a possible plasticity response after responsive neurostimulation implantation and continued stimulation.

Based on a presentation by Andrew Cole, MD, FRCP(C), Professor of Neurology, Harvard Medical School; Director, Massachusetts General Hospital Epilepsy Service; and Chief, Division of Clinical Neurophysiology Laboratory, Massachusetts General Hospital, Boston, Massachusetts.

The subject of disease-modifying agents involves the two main obstacles of cost and time. For chronic epilepsy, the cost of screening AEDs is around $10,000 for in vitro studies, $100,000 for in vivo testing, and millions of dollars for clinical trials. The evaluation time increases from days to years when moving from in vivo studies to human trials, and the number of compounds tested drops dramatically when clinical trials become involved.

Testing of medications to treat particular seizure types illustrates the promise of disease-modifying AEDs. Epileptic encephalopathies are childhood syndromes characterized by spikes, seizures, and cognitive delay/regression with possible sleep activation. They may be symptomatic or idiopathic. The number of spikes and seizures in this patient population is a risk factor for developmental effects. The expression of epilepsy in these patients may be age-dependent, and cognitive dysfunction often continues regardless of seizure activity. Earlier initiation of therapy and subsequent response may be associated with improved cognitive outcomes; these findings appear to be independent of the specific treatment chosen (eg, adrenocorticotropic hormone, vigabatrin, or dietary therapy).24 In addition, more resistant, symptomatic etiologies are related to poorer long-term cognitive outcomes. Results of studies involving mouse models of epileptic encephalopathies25 have shown that early treatment may suppress the development of spike/wave epilepsy and further cognitive dysfunction. Whether or not this holds true for pediatric populations remains nearly impossible to determine through clinical trials.

Disease modification may provide significant benefit in patients who develop post-traumatic epilepsy. Currently, prophylactic AED treatment has been shown to reduce the risk of developing late posttraumatic epilepsy. Patients with severe traumatic brain injury have a greater risk of developing seizures than do those with mild or moderate injuries—but this greater risk remains only about 15% at 30 years. The type and location of the severe traumatic brain injury and use of more advanced imaging techniques can help clinicians to determine which patients are more likely to develop epilepsy and, possibly, to participate in clinical trials of disease-modifying agents.

The same principles may help selection of trial patient populations suffering from other potentially conditions that cause seizures (eg, stroke, tumor, infection, hemorrhage). Future studies must accurately determine how many seizures occur after an event, how severe the precipitating event was, and how long-term follow-up will be accomplished. Exploration of this subject will require time, money, and, most importantly, new innovations and consensus on appropriate study designs.


  1. Fisher RS, Aevedo C, Arzimanoglou A, et al. ILAE official report: a practical clinical definition of epilepsy. Epilepsia. 2014;55:475–482.
  2. Musicco M, Beghi E, Solari A, Viani F, for the First Seizure Trial Group (FIRST Group). Treatment of first tonic-clonic seizure does not improve the prognosis of epilepsy. Neurology. 1997;49:991–998.
  3. Kwan P, Brodie MH. Early identification of refractory epilepsy. N Engl J Med. 2000;342:314–319.
  4. Perucca E. An introduction to antiepileptic drugs. Epilepsia. 2005:46(suppl 4):31–37.
  5. Marson AG, Al-Kharusi AM, Alwaidh M, et al. The SANAD study of effectiveness of carbamazepine, gabapentin, lamotrigine, oxcarbazepine, or topiramate for treatment of partial epilepsy: an unblended randomized controlled trial. Lancet. 2007:369:1000–1015.
  6. Marson AG, Al-Kharusi AM, Alwaidh M, et al. The SANAD study of effectiveness of valproate, lamotrigine, or topiramate for generalized and unclassifiable epilepsy: an unblended randomized controlled trial. Lancet. 2007;369:1016–1026.
  7. Mattson RH, Cramer JA, Collins JF, et al. Comparison of carbamazepine, phenobarbital, phenytoin, and primidone in partial and secondarily generalized tonic-clonic seizures. N Engl J Med. 1985;313:145–151.
  8. Glauser T, Ben-Menachem E, Bourgeois B, et al. Updated ILAE evidence review of antiepileptic drug efficacy and effectiveness as initial monotherapy for epileptic seizures and syndromes. Epilepsia. 2013;54:551–563.
  9. Hernández-Díaz S, Smith CR, et al; North American AED Pregnancy Registry; North American AED Pregnancy Registry. Comparative safety of antiepileptic drugs during pregnancy. Neurology. 2012;78:1692–1699.
  10. Tomson T, Battino D, Bonizzoni E, et al. Dose-dependent risk of malformations with antiepileptic drugs: an analysis of data from the EURAP epilepsy and pregnancy registry. Lancet Neurol. 2011;10:609–617.
  11. Kwan P, Brodie MJ. Effectiveness of first antiepileptic drug. Epilepsia. 2001;42:1255–1260.
  12. Brodie MJ, Perucca E, Ryvlin P, Ben-Menachem E, Meencke HJ; Levetiracetam Monotherapy Study Group. Comparison of levetiracetam and controlled-release carbamazepine in newly diagnosed epilepsy. Neurology. 2007;68:402–408.
  13. Kwan P, Arzimanoglou A, Berg AT, et al. Definition of drug resistant epilepsy: consensus proposal by the ad hoc task force of the ILAE Commission on Therapeutic Strategies. Epilepsia. 2010;51:1069–1077.
  14. Gowers WR. Epilepsy and Other Convulsive Diseases: Their Causes, Symptoms, & Treatment. London: J & A Churchill; 1881.
  15. Brodie MJ, Yuen AW. Lamotrigine substitution study: evidence for synergism with sodium valproate" 105 Study Group. Epilepsy Res. 1997;26:423–432.
  16. Stephen LJ, Kwan P, Brodie MJ. Does the cause of localisation-related epilepsy influence the response to antiepileptic drug treatment" Epilepsia. 2001;42:357–362.
  17. Chiron C, Marchand MC, Tran A, et al. Stiripentol in severe myoclonic epilepsy in infancy: a randomised placebo-controlled syndrome-dedicated trial. STICLO study group. Lancet. 2000;356:1638–1642.
  18. Brodie MJ, Rosenfeld WE, Vazquez B, et al. Rufinamide for the adjunctive treatment of partial seizures in adults and adolescents: a randomized placebo-controlled trial. Epilepsia. 2009;50:1899–1909.
  19. Kossoff EH, Cervenka MC, Henry BJ, Haney CA, Turner Z. A decade of the modified Atkins diet (2003–2013): results, insights, and future directions. Epilepsy Behav. 2013;29:437–442.
  20. VNS Therapy Programming Software, Model 250 (Version 11.0) Physician's Manual. Houston, Tex: Cyberonics, Inc; 2015.
  21. Morrell MJ; RNS System in Epilepsy Study Group. Responsive cortical stimulation for the treatment of medically intractable partial epilepsy. Neurology. 2011;77:1295–1304.
  22. Salnova V, Witt T, Worth R, et al. Long-term efficacy and safety of thalamic stimulation for drug-resistant partial epilepsy. Neurology. 2015;84:1017–1025.
  23. Kukiert A, Mella Cukiert C, Burattini JA, de Moura Lima, A. Seizure outcome after battery depletion in epileptic patients submitted to deep brain stimulation. Neuromodulation. 2015;18:439-441.
  24. Pellock JM, Hrachovy R, Shinnar S, et al. Infantile spasms: a U.S. consensus report. Epilepsia. 2010;51:2175–2189.
  25. Blumenfeld H, Klein JP, Schridde U, et al. Early treatment suppresses the development of spike-wave epilepsy in a rat model. Epilepsia. 2008;49:400–409.

Dr. Houston is a Clinical Neurophysiology Fellow at the University of Alabama at Birmingham, Birmingham, Alabama.

View/download a PDF version of this article