Abstracts

Rationale: Magnetoencephalography (MEG) becomes an arsenal of noninvasive preoperative evaluations to detect the epileptogenic zone of focal cortical dysplasia (FCD). If the epileptogenic zone is widespread or complicated, conventional equivalent current dipole (ECD) analysis could fail to delineate the entire epileptogenic zone. Gradient magnetic-field topography (GMFT) is a unique analysis to demonstrate spatiotemporal dynamics of brain surface activity without a solution of neuromagnetic inverse problem. GMFT can localize the spike activity even at the early phase of the spikes with low signal-to-noise ratio and gyral level accuracy.The aim of this study is to validate the detectability of interictal spike onset area, so called “spike onset zone”, using GMFT and to evaluate the additional value of combination of GMFT and ECD analysis. Methods: We enrolled 41 patients (2–58 years old, mean age 24.1±12.5; 26 males and 15 females) with neocortical FCD. MEG spikes were analyzed by ECD and GMFT. GMFT was applied at early onset of MEG spikes (GMFT-O). ECD was performed at the spike peak. The distribution of GMFT-O and the distribution mismatch between GMFT-O and ECD were evaluated by gyral level on 3D surface MRI to classify into concordant (C), partially concordant (PC), and discordant (D). The distributions and correlations were compared with three FCD subtypes (ILAE criteria, 2011) and surgical outcomes. Good seizure outcome groups consisted of Engel class I, and poor seizure outcome group was Engel class II–IV. Results: FCD subtypes were classified into three groups according to the ILAE criteria (2011): type I (n=17), type IIa (13), and type IIb (11). Seizure outcomes were Engel class I in 20 (48.8%), class II in 6 (14.8%), and class III in 15 (36.6%).GMFT-O of FCD type IIb distributed significantly narrower (1–2 gyri, 90.9%) than those of type IIa (3–4 gyri, 69.3%) and I (3–4 gyri, 47.1%) (p=0.039).Type IIb presented significantly more C group (7/11, 63.7%) than type IIa (1/13, 7.7%) and type I (1/17, 5.8%). PC and D were more frequently seen in type I and IIa than IIb.PC (3/11, 27.3%) and D (1/11, 9.1%) were less found in type IIb than IIa (PC, 9/13, 69.2%; D, 3/13, 23.1%) and I (PC, 10/17, 58.2%; D, n=6/17, 35.3%) (p=0.0006).In 20 patients with good seizure outcome; the complete resection (9/10, 90%) of GMFT-O was more performed than partial resection (7/23, 30.4%) and no resection (4/8, 50%) with a statistical significance (p < 0.0001); the resection of ECD did not differ among the complete resection (9/13, 69.2%), partial resection (9/19, 47.4%), and no resection (2/9, 22.2%) (p=0.0844). Conclusions: GMFT can reveal the spike onset zone in three types of FCD. The distribution of GMFT-O, and the correlation of GMFT-O and ECD reflect the characteristics of the epileptogenic zone of FCD subtypes. The complete resection of spike onset zone derived from GMFT achieved better seizure outcome than resection of ECD areas. The spike onset and peak zones interictal epileptic discharges in FCD can lead to understand the entire epileptogenic zone. Funding: This study was supported in part by a Health Labour Sciences Research Grant from the Ministry of Health, Labour and Welfare of Japan.