Leijten Frans S S, Huiskamp Geert-Jan M, Hilgersom Irene, Van Huffelen Alexander C
Department of Clinical Neurophysiology, Rudolf Magnus Institute of Neuroscience, University of Utrect, The Netherlands.
J Clin Neurophysiol. 2003 Jul-Aug;20(4):227-38. doi: 10.1097/00004691-200307000-00001.
Magnetic source imaging is claimed to have a high accuracy in epileptic focus localization and may be a guide for epilepsy surgery. Non-lesional mesiotemporal lobe epilepsy (MTLE), the most common form of epilepsy operated on, has different etiologies, which may affect the choice of surgical approach. The authors compared whole-head magnetoencephalography (MEG) with high-resolution EEG for source identification in MTLE. Nineteen patients with unilateral, nonlesional MTLE underwent a simultaneous 151-channel CTF MEG (CTF Systems, Inc., Port Coquitlam, British Columbia, Canada) and 64-channel EEG recordings with sleep induction. Three independent observers selected spikes from the EEG and MEG recordings separately. Only when there was interobserver agreement (kappa>0.4) on the presence of spikes in recordings were consensus spikes averaged. EEG and MEG equivalent current dipoles (ECD) were then integrated in the head model of the patient reconstructed from MRI. The results were compared with intraoperative electrocorticography findings. Spikes were detected in 32% of MEGs and 42% of EEGs. No patient showed MEG spikes only. Equivalent current dipole modeling correctly localized the source to the temporal lobe in four out of five MEG and three out of eight EEG recordings. MEG localized sources were more superficial and EEG localized sources were deeper. Unfortunately, basal temporal lobe areas were only partially covered by the sensor helmet of the MEG setup. Best correlation between EEG or MEG findings and electrocorticography findings was between horizontal EEG dipole orientation and prominent neocortical spiking; these patients also had a less favorable prognosis. Magnetic source imaging is currently unlikely to alter the surgical management of MTLE. The yield of spikes is too low, and ECD modeling shows only partial correlation with electrocorticography findings. Moreover, the whole-head MEG helmet provides insufficient coverage of the temporal lobe.
磁源成像在癫痫病灶定位方面据称具有较高的准确性，可能为癫痫手术提供指导。非病灶性颞叶内侧癫痫（MTLE）是最常见的接受手术治疗的癫痫类型，其病因各异，这可能会影响手术方法的选择。作者比较了全头磁脑图（MEG）与高分辨率脑电图在MTLE病灶识别中的应用。19例单侧、非病灶性MTLE患者同时接受了151通道的CTF MEG（CTF Systems公司，加拿大不列颠哥伦比亚省高贵林港）和64通道脑电图记录，并进行睡眠诱发。三名独立观察者分别从脑电图和MEG记录中选取棘波。只有当观察者之间就记录中棘波的存在达成一致意见（kappa>0.4）时，才对一致的棘波进行平均。然后将脑电图和MEG等效电流偶极子（ECD）整合到根据MRI重建的患者头部模型中。将结果与术中皮质脑电图结果进行比较。在32%的MEG和42%的脑电图中检测到棘波。没有患者仅表现出MEG棘波。等效电流偶极子建模在五分之四的MEG记录和八分之三的脑电图记录中正确地将病灶定位到颞叶。MEG定位的病灶较浅，脑电图定位的病灶较深。不幸的是，MEG设备的传感器头盔仅部分覆盖颞叶基底区域。脑电图或MEG结果与皮质脑电图结果之间的最佳相关性存在于水平脑电图偶极子方向与明显的新皮质棘波之间；这些患者的预后也较差。目前，磁源成像不太可能改变MTLE的手术管理。棘波的检出率过低，并且ECD建模仅显示与皮质脑电图结果存在部分相关性。此外，全头MEG头盔对颞叶的覆盖不足。
