Intractable epilepsy: the surgical option

Julia B. Toub, M.D.
General neurologist, epilepsy
Providence Epilepsy Services

Published March 2011

Epilepsy, defined by at least two unprovoked seizures, is a common neurological illness, affecting up to 2 percent of the population. Although epilepsy is largely controlled with antiepileptic drugs, approximately 30 to 40 percent of patients continue to have seizures despite appropriate medications.

An estimated 600,000 Americans meet the criteria for intractable epilepsy, defined as continued epileptic seizures despite adequate trials of at least two antiepileptic drugs. Admission to an epilepsy monitoring unit for continuous video EEG recording may provide valuable diagnostic information in distinguishing between epileptic and nonepileptic seizures.

Although not all patients with intractable epilepsy are eligible, surgery holds promise for many.
Pathologic substrates for pharmaco-resistant epilepsies are diverse, though a large subset of cases are associated with mesial temporal sclerosis. Other common causes for intractability include cortical dysplasias, low-grade gliomas, dysembrioplastic neuroepithelial tumors and hamartomas.

We now know that seizures arising in one region may cause irreversible damage and eventually kindle seizure activity in other regions of the brain, thereby establishing new, independent foci for epileptogenesis. As such, early and aggressive seizure management is essential.

Although not all patients with intractable epilepsy are eligible, surgery holds promise for many patients in this population. Evaluation for surgery is a multi-step process, usually beginning with continuous video EEG monitoring, through which seizures are analyzed on the basis of EEG as well as clinical semiology (how the patient behaves during the seizure).

MRI, functional tests such as PET, SPECT and fMRI, invasive EEG modalities and sophisticated surgical strategies have helped more patients than ever achieve seizure freedom. Eventually, cortical regions involved in seizure propagation can be identified and resected, ideally rendering the patient seizure-free with minimal functional deficit.

The following is a brief summary of these modalities and their application in surgical evaluation.

Continuous video EEG monitoring
Video EEG is the compulsory first step in presurgical evaluation. Patients are admitted to the hospital for characterization of their seizures. EEGs are scrutinized for overt epileptic activity and subtle patterns indicative of underlying pathology, and seizure semiology is reviewed for potential localizing or lateralizing features.

Magnetic resonance imaging
MRI has become a favorite test across specialties because it provides practitioners with valuable anatomical information. Recently, high-resolution 3-tesla magnets have allowed neurologists to see lesions not previously detected on 1.5-tesla MRI scans. For epileptologists, this translates into discovery of subtle cortical abnormities and potential sites of seizure onset. This knowledge may become especially valuable for planning invasive presurgical strategies as well as eventual resection.

Functional studies
Modalities such as positron emission tomography (PET), single photon emission computed tomography (SPECT) and functional MRI (fMRI) are key steps in presurgical evaluation. PET uses radiotracer to evaluate metabolic activity within tissues, since hypometabolic regions often coincide with foci of epileptogenicity.

SPECT also uses a radioactive substance, but contrary to PET it is a biphasic test comparing the ictal (seizure) and interictal (non-seizure) perfusion state of the brain. During ictal SPECT, radiotracer injected at seizure onset is preferentially taken up by brain regions involved in seizure propagation representing relative hyper-perfusion compared to the resting state.

Using SISCOM (subtraction ictal SPECT co-registered to MRI) technology, the interictal data is subtracted from the ictal data and fused with MRI, delineating those regions of the brain that are involved in seizure onset.

Aside from localizing the ictal onset zone, a major goal of presurgical evaluation is to delineate hemisphere dominance and eloquent cortex. Functional MRI uses noninvasive measures to detect regional changes in blood flow during specific tasks.

Though helpful in many cases, fMRI results may be equivocal, thus necessitating a more invasive test: the intracarotid sodium amobarbital procedure, or Wada test. In this test, following selective anesthetization of each hemisphere with amobarbital, the patient is asked to complete tasks designed to identify language dominance. With the advent of fMRI, Wada has fallen in popularity as it carries a risk for stroke (less than 1 percent), and results are often confounded by encephalopathy and anatomic variation in blood flow.

Neuropsychiatric evaluation
The importance of neuropsychiatric testing is often underappreciated. Preoperative assessment can strengthen hypotheses regarding neuroanatomical seizure focus. For example, a patient with left temporal lobe epilepsy might have poor verbal memory, whereas prefrontal lesions may manifest as executive dysfunction. Furthermore, testing batteries can be used to predict which neurocognitive domains are at risk for deterioration should the patient proceed with surgery.

Invasive evaluation and cortical mapping
In some patients, noninvasive modalities are insufficient for seizure localization. In others, the suspected seizure focus lies close to eloquent cortex. In these patients, a second phase of EEG evaluation becomes necessary.

Data obtained from scalp EEG, neuroimaging, functional tests and neuropsychiatric evaluation are used to design an invasive strategy. In the operating room a grid of electrodes is placed subdurally over the patient’s cortical area of interest. Deeper foci may be evaluated with depth electrodes. Continuous video EEG monitoring allows for precise seizure localization.

In patients with seizures near speech or motor areas, subdural electrodes can be electrically stimulated. This meticulously maps eloquent functions and predicts deficit should that area of cortex be removed.

Certainly, we are living in an exciting time in which technology has vastly changed the way we practice medicine. Surgical treatment of epilepsy, though not appropriate for all, holds promise for select patients in which drugs just don’t cut it.