Intraindividual analysis of antiepileptic drug effects on EEG background rhythms

Antiepileptic drug (AED) therapy with either phenytoin or carbamazepine has been associated with generalized slowing of EEG background rhythms. These effects have been seen in groups of patients undergoing D manipulation, although the background slowing has been highly variable from patient to patient. Background slowing may represent an objective physiologic measure of drug impact on cerebral function and could be useful in monitoring for AED neurotoxicity in individual patients. This would require an intraindividual analysis of AED effects on EEG background rhythms. The present study was designed to develop a methodology for intraindividual analysis of EEG background changes and to apply this methodology to patients beginning or ending chronic AED therapy. EEG recordings were obtained under controlled conditions in 31 healthy subjects and were repeated after an interval of 12-16 weeks. EEG background rhythms from each record were analyzed using the fast Fourier transform, and test-retest differences for several quantitative measures were calculated from each subject's paired recordings. EEG recordings were also obtained in 6 patients beginning or ending chronic AED therapy. Test-retest differences for each patient's quantitative EEG measures were statistically compared with the distributions of test-retest measures obtained from the healthy controls. AED therapy was associated with an increase in absolute delta and/or theta power and a slowing of the dominant posterior rhythm; however, these EEG background changes varied widely in degree from patient to patient. Intraindividual analysis revealed that 5 patients had statistically significant slowing relative to the control group on at least 1 of the 9 target quantitative EEG measures, as well as a composite measure. Statistically significant slowing was also seen as a group effect. The results suggest that among patients undergoing chronic therapy with phenytoin or carbamazepine there are widely different degrees of EEG background slowing, often significantly beyond that expected on the basis of test-retest variability. Intraindividual analysis of the EEG background may be a practical objective measure of a patients unique response to chronic AED therapy.