Drug-resistant Epilepsy

Warren T. Blume

Introduction

This chapter discusses some factors that promote antiepileptic drug-resistant epilepsy in 20% to 30% of patients who have seizure disorders.

Accurate and prompt recognition of an epileptic condition that is likely to be intractable to medical and psychological therapy will guide management and the tendering of a prognosis to family and close associates. Unexpected intractability raises the possibility of a focal or diffuse progressive disease, discussed in Chapters 252 and 264.

After an initial section on definitions of intractability, principal aspects pertaining to focal and generalized epilepsies will follow. Interposed between focal and generalized sections, and applying to each, are brief discussions of propagation, genetics, and pharmacoresistance.

Defining Intractability

Paradoxically, defining “intractability” is probably more difficult than discerning which patients have or will have therapy-resistant seizures, as the following considerations illustrate.

Medical Treatment Failure

Persistent seizures despite about two or three appropriately selected and prescribed antiepileptic drugs constitutes medical failure, based upon the sharply declining probability that subsequent drugs will succeed.²^,⁷^,²⁶^,³¹^,⁵⁷ Accurate classification of the seizure disorder through ictal semiology, electroencephalography (EEG) and presumed etiology is prerequisite for antiepileptic drug (AED) selection and sequencing.

Importantly, seizure freedom, only obtained by overt or covert AED toxicity, constitutes intractability. Covert toxicity occurs when polypharmacy is gradually attained or augmented (e.g., with two or more drugs in the “therapeutic range”). Patients, associates, and the treating physician may all be oblivious to this circumstance.

Seizure Occurrence

Minimum seizure frequency is one approach to this component of definition.²⁶^,³¹ Does one count seizure number or number of seizure days per month or year? Seizure type and circumstance must be considered: A single tonic–clonic seizure in a truck driver will have more impact than rare, brief absences in a school child. The opposite approach is establishing the minimum duration of remission needed to not qualify as intractable.² With either or both criteria, patients could be classified as always, usually, seldom, or never intractable for each seizure type.

Time Dimension

To avoid hastily labeling as intractable certain patients having seizures with an acute illness or other circumstance, measures reflecting frequency with time have value: (a) minimum time that one is not free of seizures (e.g., 6 months or 12 months), (b) outcome at a specific point in time (e.g., failure to be in 6-month remission 2 years after diagnosis), and (c) minimum seizure frequency at any point in the disorder.⁸

Strict Versus Loose

A strict definition would have high specificity but may miss some patients who should be managed and followed as intractable. A loose definition would unfairly label, depress, and overinvestigate some patients. As the use of this definition will determine the stringency required, the concept of an intractability scale would be useful.⁸⁶ For example, a loose definition—or low point on a scale—would be appropriate for referral to an epileptologist, while a strict definition may be required for invasive investigations or surgery.

Children

Principles underlying any definition of intractability may closely resemble those generally applicable, except that age will play a relatively important role. An earlier age of onset correlates with a higher prevalence of intractability.⁶^,⁴⁴ However, studies have shown an approximately similar prevalence of intractability in children as in adults, possibly as a higher proportion of benign syndromes that begin in childhood statistically counterbalances the several malignant syndromes of childhood onset (see further).

Intractability may be more difficult to predict in the pediatric age group. Berg et al.⁷ found that outcome could not be predicted in almost 40% of children after 2 years of follow-up. Camfield and Camfield²⁰ found similar predictability in patients with dyscognitive seizures. Additionally, Berg et al.⁷ reported ultimate remission in 14% of previously intractable patients. See Farrell et al.³² for further discussion.

Extraneous Factors

Optimal application of any definition or set of criteria requires reliable data. Thus, an accurate diagnosis is essential, bearing in mind that many intermittent central nervous system events
represent nonepileptic disorders (e.g., syncope, migraine, pseudoseizures, and diurnal drowsy episodes when sleep deprived). This principle applies to both initial and follow-up information.

Additional causes of misleading data are AEDs inappropriate for a given seizure disorder, the wrong dose of a correct medication, or suboptimal compliance with the prescribed regimen. Premature abandonment of a useful drug may occur because of excessive haste to obtain seizure control or through overoptimistic therapeutic goals for disorders known to be intractable.

Table 1 Some Aspects Auguring Intractability

Focal epilepsies
   Region
      Temporal
      Occipital
      Primary motor cortex
      Supplementary sensory motor area
Etiology
   Mesial temporal sclerosis
   Cortical dysplasia
   Hemorrhagic lesions
   Multifocal epileptogenesis
   Progressive lesions
Generalized epilepsies
   Onset in infancy or early childhood
   High initial seizure frequency
   Failure of initial appropriate AED
   EEG
      Abundant multifocal or bisynchronous spikes
      Abnormal “background” activity
   Progressive disorders

AED, antiepileptic drug; EEG, electroencephalography.

Stress increases seizure frequency.³⁷^,⁶⁷ Stress is often associated with depression, which disrupts sleep; this, in turn, lowers a seizure threshold.

Intractable Focal Epilepsies

Several interdependent factors may render a focus of epileptogenesis resistant to AEDs. This section discusses region of ictal onset, etiology, and multifocality of epileptogenesis (Table 1).

Brain Region

Temporal

The mesial temporal lobe is probably the most epileptogenic of brain regions. Hauser and Kurland,⁴⁶ in an epidemiologic study, found temporal lobe epilepsy (TLE) more prevalent (1.7 per 1,000) than all other focal epilepsies combined. Ninety percent of temporal lobe seizures begin in the mesial (i.e., limbic) components, as invasive EEG recordings have disclosed.¹¹^,⁸⁴ Such recordings have shown that temporal lobe seizures most commonly begin regionally within the mesial temporal lobe, involving several components simultaneously. The less common focal onsets arise more often in the hippocampus than in the amygdala.²⁹^,⁷⁷^,⁸³ Paradoxically, kindling, an important experimental model of epileptogenesis, is most easily elicited by stimulation of limbic structures,³⁸ especially the amygdala.¹⁹

Occipital

Ebersole and Chatt²⁸ found layer four of the striate cortex to be highly epileptogenic from effects of discrete penicillin applications to each layer. Similarly, Cain¹⁸ readily produced kindling by electrical stimulation of the deep layers of the occipital cortex, whereas stimulation of superficial layers failed to evoke kindling.

Motor Cortex and Supplementary Sensory Motor Area

The face and hand area of the motor cortex is also a common site of seizure origin.³³^,⁹³ A fourth region often involved in intractable seizures is the supplementary sensory motor area.⁷²^,⁷⁴ Seizures from this region may be intractable because of its abundant efferents to several levels of the motor system.¹⁰^,⁶⁴^,⁹²

Etiology

Focal seizure etiology influences intractability at least as strongly as location. Mesial temporal sclerosis (MTS), cortical dysplasia, and hemorrhagic lesions are examples.

Mesial Temporal Sclerosis

Unfortunately, an unambiguous explanation of MTS epileptogenesis has not emerged from experimental and clinical data as causative, compensatory, and associative factors are intertwined. Although several basic and clinical studies have found data consistent with an increased or maintained γ-aminobutyric acid (GABA) inhibitory system,¹⁷^,³⁰ such inhibition has been shown to be fragile⁹⁵ or even excitatory in certain circumstances.³

Spontaneous epileptic activity occurs in the entorhinal cortex in humans with epilepsy.⁵ Collins et al.²³ found that the dentate gyrus only initially restricted propagation of penicillin-evoked seizures in rat entorhinal cortex; as this inhibition was overcome, seizure activity engulfed the hippocampus. Leung and Wu⁵⁹ described an increase in excitation of the rat entorhinal cortex–dentate gyrus pathway in kainic acid–treated rats and in CA1 of the hippocampus from CA3 stimulation. Vulnerability of CA3 stratum lucidum interneurons has also been found experimentally,⁶⁵ which may lead to an extensive remodeling of mossy fibers in CA3. This may shift the dentate granule–CA3 pathway from feed-forward inhibition to direct excitation, leading to excessive recruitment of CA3 pyramidal cells in epileptic activity.⁵⁸ Ratte and Lacaille⁷⁸ documented reduction of GABAergic cells in the stratum oriens and alveus in the kainic acid model, increasing the ratio of excitatory to inhibitory synapses in the stratum lacunosum moleculare. Thus, CA1 becomes hyperexcitable despite loss of a large proportion of Schaffer collateral afferents.

Several studies have disclosed enhanced excitation in the epileptogenic mesial temporal region. Leung and Wu⁵⁹ described opening of sodium channels with less depolarization producing larger sodium currents. Additionally, an increased pool of releasable glutamate and increased N-methyl-D-aspartate (NMDA) receptor activation in MTS has been found.³⁹^,⁶⁰

Only gold members can continue reading. Log In or Register to continue