The goal of the surgical evaluation is the localization of the seizure onset and the determination as to whether the focus can be resected safely. Because no single diagnostic test is capable of defining the spatial extent of the epileptogenic zone, the key to the preoperative evaluation is obtaining concordant data that defines and maps out the specific region likely to be the etiology of the seizures. This often requires an extensive evaluation that, in most major institutions performing epilepsy surgery, utilizes a multidisciplinary team approach that includes specialists in neurology, neurosurgery, neurophysiology, neuroradiology, and neuropsychology. The decision to proceed with a further evaluation and/or surgical intervention is made after compilation of the data from each phase of the work-up and comprehensive review of the findings by the epilepsy group.
For convention, the evaluative and treatment process is divided into four phases. The first phase is the initial, noninvasive work-up that has its goal the definition of the seizure syndrome and potentially ensuring that the medical therapy is maximized. The second phase or invasive monitoring is used to further lateralize and localize the seizure focus that remains ill defined by the noninvasive studies or used to determine the location of the seizure focus relative to eloquent cortex. The third phase or the decision for a surgical intervention for seizure control follows comprehensive review of the data, agreement by the epilepsy group that indeed there is a concordance of findings, and that surgery is the best option either curative or palliative. The fourth phase or the postoperative assessment is routinely performed 6-24 months after the surgical intervention for a final assessment of the outcome. The concepts of the assessments or surgical evaluation do not differ significantly between adults and children, though specific subtleties do exist.
The first phase in the initial, noninvasive evaluation routinely includes an extensive history, including pre- and postnatal events, seizure history including the semiology or specific characteristics of the onset and manifestations of the seizures, a neurologic examination, the electrophysiology through multiple electroencephalography (EEG) exams that record interictal and ictal events, and neuroimaging. The neuroimaging component of the evaluation, specifically MR imaging, has become absolutely indispensable in delineating anatomic abnormalities that may be the etiology of the seizure focus. These include focal and diffuse cerebral pathologies, such as neoplastic lesions, vascular malformations, cytoarchitectural lesions, such as neuronal heterotopias, gyral anomalies, and other abnormalities, though the spectrum of pathologic entities that can cause seizures is innumerable. With respect specifically to the temporal lobe and the amygdalohippocampal complex, MTS is frequently imaged as tissue signal changes and volumetric differences between the hippocampi and often confirms the EEG findings as to the underlying seizure focus. The imaging sequences of MR that have been found to be most useful in the epilepsy evaluation include the long TR, and the fluid attenuation inversion recovery (FLAIR) sequences. These improvements in technique and technology have improved the identification of subtle medial temporal abnormalities in addition to the gross cortical and lesional findings frequently seen in these patients.
Further advances in neuroimaging have added to the armamentarium of the epilepsy team. With the advent and increased utilization of functional imaging with SPECT or PET in conjunction with MR imaging, there has been a significantly diminished need for invasive monitoring. When the MR and/ or functional imaging correlate with the electrophysiology and other evaluative measures, there is adequate information to recommend and proceed with a surgical intervention. If there is no concordance of data or it is contradictory, further or secondary testing during the Phase I evaluation may be required. A hospital admission for longer continuous scalp EEG/video-telemetry may help to better define the seizure frequency and semiology, particularly if the seizures are predominantly nocturnal or ill-defined on routine EEG and is necessary if surgery is being considered. Neuropsychological testing can assess baseline performance, the level of development, deficits in verbal and nonverbal communication, and psychosocial adjustment which may help to elucidate the extent of premorbid damage and to predict outcome following epilepsy surgery. Intracarotid injection of amytal, known as the WADA test, induces a transient inactivation of targeted brain regions, which produces a "reversible resection" allowing assessment of the potential effects of surgery on specific brain regions. The Wada test has long been used for determining hemispheric speech dominance and memory representation and avoiding resections of eloquent areas though it is not perfect. Because this type of testing requires cooperation through a long an arduous exam, Wada testing is more difficult in young children, and is usually reserved for patients in whom language or memory lateralization is critical for surgical decision making. In the future, functional MR may supplant the need for the angiogram and Wada testing. Testing of language areas and speech dominance is being utilized in some specialized centers even in young children.
When the non-invasive methods of evaluation are insufficient to determine the origins of the epileptogenic focus or when the epileptogenic zone must be defined with high precision in relation to nearby eloquent cortex, invasive EEG monitoring is often recommended. Indications for long-term invasive monitoring include possible multiple seizure foci, unclear lateralization and localization, and extratemporal foci. The need for extraoperative monitoring and mapping is more likely in a child with extratemporal epilepsy than with temporal lobe epilepsy.
Intraoperative electroencephalography (EcoG) and cortical mapping in the awake patient can be used to stimulate and to record from the cerebral cortex during surgery in order to reduce the need for chronic invasive monitoring. Unfortunately, intraoperative ECoG is a limited option in children. Many children, particularly the preadolescent child, cannot tolerate surgery under local anesthesia and offer the cooperation necessary for definitive mapping of the seizure focus and eloquent areas. Other reasons for its limited usefulness include:
- the information obtained from intraoperative monitoring most often is interictal and obtained in an abnormally stressful environment;
- anesthetic agents may reduce seizure activity by altering the thresholds of both the afterdischarges and the motor responses; and
- the extended operative time required for cortical mapping.
For these reasons, extraoperative monitoring provides a broader range of data regarding the seizure focus, including ictal and interictal data of the typical habitual seizures for the individual patient. Also using chronically implanted electrodes, data regarding the relationship of the seizure focus to the surrounding tissue that may be functionally significant can also be obtained. The types of invasive monitoring consist of either DE or surface electrodes.
DE are stereotactically placed intracerebral contacts that predominantly have been used to define temporal and inferior frontal lobe pathology. They are particularly useful when the seizures are thought to be of medial temporal origin, but lateralization is uncertain. For the most part, DE have been used in adults, although they can be inserted safely in children. Their advantages over surface electrodes are that DE have a higher sensitivity for lateralization, and can be inserted through a twist drill hole, thus obviating a craniotomy. The main disadvantage is that the electrophysiologic information obtained is limited to the local regions of the tissue in which the contacts are placed. As a result, the ability of DE to study epilepsy outside of the medial temporal lobe is suboptimal. Additionally, if the parietooccipital approach is used to place the electrodes, a significant distance of brain parenchyma must be traversed for the contacts to lie within the long axis of the hippocampal complex, which is believed to contribute to their slightly increased morbidity due to hemorrhage.
Because many epileptic children, especially those <12 years of age, have epileptic foci that are cortical and extratemporal, DE monitoring is often suboptimal. Because these children frequently require assessment of the lateral cortical surface and surrounding regions, including the medial temporal structures, cortical surface electrodes such as grid and strip electrodes are frequently recommended, even in children as young as 1 year of age. Extraoperative monitoring with cortical strip and grid electrodes has the advantage of allowing for an extended recording of seizure events, as well as providing the capability for cortical stimulations to localize eloquent areas. While a waiting period is suggested in some cases, a focused surgical resection of the epileptic zone can be performed so as to preserve eloquent cortex at the completion of the invasive evaluation.