Abstracts

Slow activity in the frequency range of 0.5- 4 Hz has been described in the MEG of patients with brain tumors and ischemic lesions. DC MEG recordings have also been performed in migraine patients. To date there is no information in the literature on ultra-slow activity (0.1-0.5 Hz) in epilepsy patients. This study intended to explore to what extent these frequencies can be demonstrated and their relationship to ictal activity and the EEG. The data were obtained on a Neuromag system. In the early cases 122 gradiometer sensors were available. In later ones 204 gradiometer and 102 magnetometers were used and the EEG was co-registered on 30 or 60 channels. The data were acquired with a bandpass from 0.1 - 90 or 0.1-120 Hz. Offline data analysis was performed with the BESA software. Ictal recordings were available in three patients and interictal in ten. Background ultra-slow activity, as defined above, was present in all recordings with an amplitude ranging between 6-113 fT (corresponding EEG amplitude was 5-16 [micro]V). There was also intermittent focal increase in amplitude in various brain regions but these did not always correlate with the areas of interictal spike activity. In 9 patients continuous rhythmic activity of 4-5 second duration was present in the occipital or frontal areas which tended to wax and wane suggesting a biologic phenomenon but its nature is unclear at this time. Intermittent burst activity into the pT range was observed in 10 patients. These bursts were also visible in the usually sampled frequency range of 1-90 Hz but much shorter in duration. Relationships to seizure onset could not be observed and the mentioned MEG bursts had no consistent EEG correlates. Since MEG is less contaminated by skin currents ultra-slow activity may be more reliably recorded by MEG than EEG. These currents exist as background activity but can show also widespread burst discharges in epilepsy patients. The relationship of focal and paroxysmal ultra-slow activity to other parameters needs further investigation as well as the nature of continuous rhythmic activity, which may be related to respiration. Further studies, which include the DC range, are indicated since they may demonstrate the internal milieu of the brain involved in epileptogenesis.