Objective   To compare the accuracy between magnetoencephalography (MEG) and video-electroencephalography (V-EEG)

  single photon emission computed tomography (SPECT) or positron emission tomography (PET) for preoperative location of epileptogenic focus in patients with refractory epilepsy.    Methods   MEG and V-EEG were performed on 73 patients with refractory epilepsy to determinate the epileptogenic focuses for operation. Fifty-two of them were examined by V-EEG plus sphenoidal electrode

19 by PET

and 48 by SPECT at the same time. Epileptogenic focus resection were performed under the direction of intra-operative electrocorticography (ECoG) in all of these 73 cases. The accuracy of pre- and intra-operative techniques for detecting epileptogenic focus were analyzed.    Results   Out of the 73 cases

epileptogenic focus were detected by preoperative MEG in 66 cases

with an accuracy rate of 90.41%

by V-EEG in 44 cases (60.27%). Among the 52 cases detected by preoperative V-EEG plus sphenoidal electrode

the accuracy was 65.39% (34 cases)

and for the 19 cases by PET and 48 cases by SPECT

it was 52.63%  (10 cases) and 47.92% (23 cases) respectively (MEG compared with V-EEG

V-EEG plus sphenoidal electrode

PET and SPECT respectively

P0.05). For the 14 cases with temporal epilepsy due to  hippocampal sclerosis

only 8 (57.14%) was detected by MEG preoperatively

but 12 cases (85.71%) was found when V-EEG plus sphenoidal electrode recording was performed (P0.05).    Conclusion   MEG improves the accuracy of preoperative location of epileptogenic focus in refractory epilepsy surgery; but for the patients with temporal epilepsy

V-EEG plus sphenoidal electrode recording maybe a better choice.