The Value of MEG in Evaluating Patients for Epilepsy Surgery
Abstracts & Commentary
Sources: Lamusuo S, et al. [18-F]FDG-PET and whole-scalp MEG localization of epileptogenic cortex. Epilepsia 1999; 40(7):921-930; Wheless JW, et al. A comparison of magnetoencephalography, MRI, and V-EEG in patients evaluated for epilepsy surgery. Epilepsia 1999;40(7):931-941.
Magnetoencephalography (meg) noninva-sively records the magnetic fields produced by electrical brain activity and offers a noninvasive method to localize brain function. Previously almost exclusively a research tool, MEG is increasingly used clinically to localize normal and abnormal cortical activity prior to brain surgery. One of the earliest clinical applications of MEG was to localize the origin of epileptic spikes. Unlike the electric fields recorded by conventional electroencephalography (EEG), neuromagnetic fields detected by MEG pass through the skull and scalp undistorted. MEG can localize focal neural activity with a spatial resolution of a few millimeters. The early MEG instruments, however, had few detectors and required repeated measurements for a complete examination. Early MEG studies were time consuming and impractical for routine clinical use. The advent of new MEG systems able to record simultaneously from a large area of scalp, or even from the entire scalp, reduces the time necessary for examination and increases the efficiency of MEG. The clinical use of current MEG systems in the presurgical evaluation of patients with medically intractable epilepsy is the subject of two recent studies by Lamusuo and colleagues and Wheless and colleagues.
Wheless et al prospectively compared surgical outcome in 58 consecutive patients with complex-partial seizures with the location of epileptogenic regions predicted by MEG, MRI, surface EEG, and subdural EEG. For each method, Wheless et al applied a "concordance" scale that characterized the overlap of the predicted epileptogenic region(s) by each method and the region resected at surgery. Surgical outcome was used to rate each method by whether a predicted epileptogenic region hit, partially hit, or missed the clinical goal of seizure relief. Overall, taking both patients with temporal and extratemporal resections, ictal subdural EEG hit in 69% of patients, and MEG hit in 52%. By comparison, interictal subdural EEG (48% hits), ictal and interictal surface EEG (32% and 44% hits, respectively), and MRI (48% hits) did no better than MEG. There was no statistically significant difference between methods. In the subgroup of patients with extratemporal resections, MEG achieved 44% hits, compared to 81% and 75% hits for ictal and interictal subdural EEG, a statistically significant difference in favor of subdural EEG. Once again, the surface EEG and MRI did no better than MEG. In summary, in this series of 58 patients with complex-partial seizures, a single MEG study identified an epileptogenic region of cortex as well as continous surface EEG recordings that required inpatient hospitalization.
Currently, the routine presurgical evaluation of epileptics usually includes high-resolution brain MRI, ictal surface EEG, and, frequently, positron emission tomography (PET). It is possible that a combination of noninvasive methods, such as MEG, MRI, and PET, can reduce or eliminate the need for prolonged inpatient surface EEG monitoring. Lamusuo et al studied nine epileptic patients who were undergoing evaluation for epilepsy surgery and who, by surface EEG, had either extratemporal epilepsy or poorly localized seizure onset. They compared the information obtained by MEG, [18F]-fluorodeoxyglucose PET (FDG-PET), and subdural EEG, and surgical outcome in operated patients’ recordings to ascertain the consistency in information provided by each method. Of nine patients, six underwent surgery. Outcome was best in five surgical patients where MEG and FDG-PET results agreed. These patients either became seizure free (n = 2), had rare seizures (n = 1), or had a more than 80% reduction in seizures (n = 2). In one surgical patient, MEG and FDG-PET results disagreed. This patient received no benefit from surgery. Of the patients who did not have surgery, two had concordant MEG and FDG-PET findings that identified regions of eloquent cortex that were unresectable. One patient had bilateral epileptic foci on MEG and subdural EEG, and did not receive surgery. MEG and FDG-PET, however, did not identify the same regions as subdural EEG in every instance. In one patient, MEG and FDG-PET did not detect an extratemporal focus seen by subdural EEG, while in another patient, MEG and FDG-PET identified an extratemporal focus that was not seen on subdural EEG. In short, this study by Lamusuo et al offers a case series in which best results following epilepsy surgery were achieved in cases where MEG and FDG-PET data agreed.
Commentary
These studies illustrate the imperfect state of presurgical assessment for epilepsy surgery. Even invasive intracranial monitoring predicted outcome (hits) in only 69% of patients. For this reason, the assessment for epilepsy surgery relies on various methods of assessment. The data presented by Wheless et al include only those patients who had surgery, so no statement can be made of the relative value of MEG in predicting which patients would not benefit from surgery. Nevertheless, these results place MEG in a similar predictive category as ictal and interictal EEG monitoring, which is the mainstay of the initial presurgical evaluation. The other major diagnostic modality used for presurgical evaluation is FDG-PET. To this end, the study by Lamusuo et al, though limited by the small number of patients, is both encouraging and cautionary. Combining FDG-PET with MEG evaluation holds the promise of identifying those patients who may benefit from epilepsy surgery, or at least from intracranial EEG monitoring. However, Lamusuo et al present evidence that MEG and FDG-PET may not detect regions of seizure onset seen by subdural EEG, and may also detect regions not seen on SD-EEG. The importance of this underinclusion or overinclusion of epileptogenic foci can only be determined through correlation with surgical outcome. To this end, there is a need for the clinical study of sufficient patients with FDG-PET and MEG to determine if this combination of noninvasive tests successfully predicts clinical outcome better than current methods. The cost of MEG is another factor that will determine its clinical use. At present, the fee for a study ranges from $1000-5000, and the cost of an MEG instrument is similar to an MRI machine. If, as these studies hint, an MEG study is clinically equivalent to inpatient surface EEG monitoring, the cost of one MEG study compares favorably with the cost of hospitalizing a patient for continuous EEG monitoring. Until the clinical applications of MEG expand, however, the high installation cost of MEG will restrict the technology to research centers and epilepsy referral centers that handle a large number of patients. —fl
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content. TESTING
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content