Detection of pathological high-frequency oscillations in refractory epilepsy patients undergoing simultaneous stereo-electroencephalography and magnetoencephalography. (April 2023)

• Record Type: Journal Article
• Authors: Vasilica, Anca-Mihaela; Litvak, Vladimir; Cao, Chunyan; Walker, Matthew; Vivekananda, Umesh
• Is Part Of: Seizure. Volume 107(2023)
• Page Start: 81
• Page End: 90
• Publication Date: 2023-04
• Subjects: Stereo-electroencephalography -- Magnetoencephalography -- High-frequency oscillations -- Drug-resistant epilepsy -- Epileptogenic zone; Epilepsy -- Periodicals; Epilepsy -- Periodicals; Seizures -- Periodicals; Épilepsie -- Périodiques; Electronic journals; Electronic journals; 616.853
• DOI: 10.1016/j.seizure.2023.03.015
• Languages: English
• ISSNs: 1059-1311

Abstract

Abstract: Background: Stereo-electroencephalography (SEEG) and magnetoencephalography (MEG) have generally been used independently as part of the pre-surgical evaluation of drug-resistant epilepsy (DRE) patients. However, the possibility of simultaneously employing these recording techniques to determine whether MEG has the potential of offering the same information as SEEG less invasively, or whether it could offer a greater spatial indication of the epileptogenic zone (EZ) to aid surgical planning, has not been previously evaluated. Methods: Data from 24 paediatric and adult DRE patients, undergoing simultaneous SEEG and MEG as part of their pre-surgical evaluation, was analysed employing manual and automated high-frequency oscillations (HFOs) detection, and spectral and source localisation analyses. Results: Twelve patients (50%) were included in the analysis (4 males; mean age=25.08 years) and showed interictal SEEG and MEG HFOs. HFOs detection was concordant between the two recording modalities, but SEEG displayed higher ability of differentiating between deep and superficial epileptogenic sources. Automated HFO detector in MEG recordings was validated against the manual MEG detection method. Spectral analysis revealed that SEEG and MEG detect distinct epileptic events. The EZ was well correlated with the simultaneously recorded data in 50% patients, while 25% patients displayed poor correlation or discordance. Conclusion: MEG recordings can detect HFOs, and simultaneous use of SEEG and MEG HFO identification facilitates EZ localisation during the presurgical planning stage for DRE patients. Further studies are necessary to validate these findings and support the translation of automated HFO detectors into routine clinical practice.