[Magnetoencephalographic studies on spike foci using a 37-channel biomagnetometer system]

We report the results of a clinical trial of Magnetoencephalography (MEG) on spike foci in patients with epilepsy, which was performed from December 1990 to June 1991 at The University of Tokyo Hospital. Fifty patients with focal epilepsy; 26 primary epilepsy, 24 secondary epilepsy (7 brain tumor, 4 arteriovenous malformation, 4 encephalitis, 3 porencephaly, 2 arachnoid cyst, 1 brain abscess, 1 hemimegaloencephaly, 1 Lance-Adams syndrome, 1 hygroma), and ten normal subjects were enrolled in this study. MEG data were recorded using a 37-channel biomagnetometer system SMI-1001 (BTi Magnes, Biomagnetic Technologies, Inc., San Diego). A simultaneous 19-channel EEG recording with linked-ear reference was also obtained. The overall study was completed safely and none of the normal subjects showed abnormal paroxysmal MEG activity. Two patients showed interictal EEG spikes which would not have been noticed without first noting the presence of corresponding prominent MEG spikes. On the whole, the MEG signal seemed to have a wider frequency bandwidth than EEG. In most cases, the source localization predicted by MEG corresponded well with the EEG findings. The relative accuracy of MEG spike source localization was estimated to be within a cubic centimeter from the cases which showed tightly clustered localization of individual spikes. High-pass filtering reduced interference by superimposed slow wave activity, thereby improving the localization of spike sources. These results demonstrate that 37-channel biomagnetometer system could be a useful tool for analyzing epileptic spike sources.