[Occipital spectral EEG parameters in newly diagnosed, untreated patients with epilepsy]

Minor spectral EEG alterations hidden to the naked eye may be of interest in the field of epileptology, cognitive performances, and drug effects. In order to introduce new scientific results of brain wave research into the clinical field of epilepsy- and drug-related cognitive problems, a normative quantitative EEG database for epilepsy was constructed.

171 newly referred, five to 50 years old patients with untreated "new" epilepsy (that is, clinical, EEG, MRI investigations had been done in 24 months after the first unprovoked seizure) were collected. EEG was recorded with closed eyes, in the waking-relaxed state. Effects that are known to influence EEG spectra (nearby seizures, drugs, etc.) were excluded as far as possible. A total of two minutes of waking-resting EEG activity was chosen for spectral analysis. Fast Fourier transformation of the selected samples were calculated resulting in absolute power, percent power and mean alpha frequency (AA, RA, and AMF respectively) for the right and left occipital derivations. For each patient (and also for 37 healthy controls), the deviation of the individual values from the age-adjusted normative mean was expressed in Z-score. Main diagnostic epilepsy categories were compared to the control group as well as to each other. In addition, effects of MRI-defined cerebral lesions and interictal spiking on spectral EEG parameters were investigated.

All group averages were within the 95 per cent confidence interval. Overwhelming majority of the individual data fell within a 3Z range. Statistically significant differences were found for AA and RA, but seldom for AMF. Right and left alpha-parameters were surprisingly symmetrical in all groups. The main difference between epilepsy groups and controls was less AA and RA power in the epilepsy groups. MRI-defined lesions and interictal epileptiform activity did not significantly influence EEG spectral variables.

These results might serve as reference data and might help planning of further quantitative EEG studies in the triangle of epilepsy, cognitive problems, and drug effects.