Assistant Professor of Clinical Neurology, Weill Cornell Medical College
Dr. Pargeon reports no financial relationships relevant to this study.
SYNOPSIS: In a review of 377 magnetoencephalography (MEG) studies in epilepsy patients undergoing presurgical workup, 44 patients were found to have one or more seizures during routine recordings, lasting up to a mean of 51.2 minutes. Ictal MEG provided unique localizing data in about one-third of patients. For patients with frequent seizures or reliably induced seizures, MEG may be a useful supplemental tool for medically refractory epilepsy patients undergoing presurgical evaluation.
SOURCE: Alkawadri R, Burgess RC, Kakisaka Y, et al. Assessment of the utility of ictal magnetoencephalography in the localization of the epileptic seizure onset zone. JAMA Neurol 2018;75:1264-1272.
Magnetoencephalography (MEG) is being used more widely for evaluating patients considering epilepsy surgery, although its availability is limited. It can add additional information for guiding phase II implantation, particularly in patients who are nonlesional on imaging or in whom scalp electroencephalogram (EEG) is nonlocalizing. MEG is noninvasive with good temporal and spatial resolution.1,2 It has several other advantages over scalp EEG, including the fact that it requires less cortical spread for detection of spike activity (3-4 cm2 vs. 6-20 cm2 for EEG), lacks distortion by intervening tissues (skull, skin, cerebrospinal fluid), and provides better spatial resolution (2-3 mm vs. 7-10 mm for EEG).2 Since MEG records tangential dipoles, it is especially useful for identifying epileptic foci associated with focal cortical dysplasias, which often are located at the bottom of deep sulci and are tangential to the surface.2 Lastly, MEG also can be used for functional mapping of eloquent cortex,2 either in lieu of or as an adjunct to functional magnetic resonance imaging (MRI) or Wada testing.
Alkawadri et al retrospectively reviewed 377 consecutive MEG studies at a single tertiary care center (Cleveland Clinic) in patients with medically refractory epilepsy undergoing presurgical evaluation from March 2008 to February 2012 to identify patients experiencing epileptic seizures during their studies. Typically, MEG was done in patients with discordant data or with non-localizing EEGs. Recordings lasted an average 51.2 minutes, using a whole-head, 306-channel MEG system, along with 21 channels of scalp EEG. Dipoles were coregistered with the patient’s brain MRI (magnetic source imaging, MSI). Ictal onset was defined as the initial period of “evolving rhythmic oscillations temporally related to the clinical and EEG onset of the patient’s typical seizures.”1 Data were reviewed by two epileptologists with expertise in MEG interpretation.
Forty-four patients had at least one seizure during their routine MEG: 25 patients (57%) had one seizure; six patients (14%) had two seizures; seven patients (16%) had three to 20 seizures; five patients (11%) had more than 20 seizures; and one patient (2%) presented for partial status epilepticus localization. The mean age of patients with seizures was 19.3 years, and they had a baseline seizure frequency of 182 seizures per month (compared to 98.7 seizures per month for excluded patients).1
Compared to scalp EEG, MEG tended to show more focal onsets in 26 patients (59%), as compared to 17 patients (39%) on scalp EEG. In fact, MEG provided unique localizing findings not appreciated on simultaneous scalp EEG in 16 patients (36%), including three patients (7%) with simple partial seizures with MEG changes without scalp EEG correlates, six patients (14%) with either nonlocalizable or generalized EEG changes, and four patients (9%) with discordant localization compared to EEG. In the latter four patients, intracranial recordings later confirmed the MEG localization. Dipole fitting was possible in 80% of patients (n = 35) with interictal discharges and 66% (n = 29) with ictal discharges. There were eight patients in whom ictal MEG provided unique findings without interictal EEG findings, seven of whom had no MRI findings, and five of whom had non-localizing findings on video EEG.
When presurgical data were reviewed in surgical management conference, epilepsy was localized or lateralized in 31 patients (70%) based on expert consensus and was nonlocalizable in the remaining 13 patients (30%). However, ictal MEG provided further localizing data in five of these 13 patients (38%). Eleven patients (25%) underwent phase II monitoring. For eight of these patients, seizure onset could be localized with intracranial EEG, and MEG dipole analysis was possible, with ictal MEG dipoles correlating to the lobe of onset in seven of eight patients (88%).
The pivotal finding is that ictal MEG provided unique localizing data for about one-third of patients who otherwise were difficult to localize with scalp EEG and other modalities. Per the authors, ictal events can be recorded in a “substantial” number of patients undergoing routine MEG, with 12% of their patients experiencing at least one event during a one-hour recording. However, the average number of seizures per month in patients with an “ictal” MEG was 182, equating to about six seizures per day. Thus, the chance of capturing an ictal event likely would be low for patients who do not experience daily seizures, unless events can be induced reliably. Also, although MEG is not sensitive to intervening tissues, movement artifact during seizures, particularly head movements, must be suppressed. Alkawadri et al recommended not only considering the incorporation of ictal MEG into patients’ presurgical analysis, but also timing ictal MEG recordings to be done when patients are admitted to the epilepsy monitoring units when antiepileptic drugs already are reduced. However, at most centers, access to MEG is extremely limited, with very few hospitals having a MEG on site. Overall, in the hands of a capable interpreter and in a patient with frequent seizures, ictal MEG may be another tool for better localizing the seizure onset zone.
- Alkawadri R, Burgess RC, Kakisaka Y, et al. Assessment of the utility of ictal magnetoencephalography in the localization of the epileptic seizure onset zone. JAMA Neurol 2018;75:1264-1272.
- Singh SP. Magnetoencephalography: Basic principles. Ann Indian Acad Neurol 2014;17(Suppl 1):S107-S112.