RESUMO
PURPOSE: Antiepileptic drugs (AEDs) can be associated with adverse neurologic effects including cognitive dysfunction. Objective methods for recognizing AED effects on the brain could be valuable for long-term management. We compared quantitative EEG measures and cognitive tests in a group of patients beginning or ending AED therapy. METHODS: Subjects included 20 patients beginning AED therapy (AEDon), 12 patients stopping AED therapy (AEDoff), 33 patient controls receiving stable AED therapy (AEDco), and 73 healthy controls (Nco). All subjects underwent structured EEG recording and a cognitive test battery before change in AED dose and again 12-16 weeks later, >or=4 weeks after the last dose change. Four occipital EEG measures (peak frequency, median frequency, relative theta and delta power) were analyzed. Cognitive test changes were scored by using test-retest regression equations based on the Nco subjects. Wilcoxon tests were used for two-group comparisons. RESULTS: AEDons had a significant decrease, and AEDoffs, a significant increase in the peak frequency of the EEG rhythm, as compared with controls. Results for median frequency and theta power were similar. Change in the EEG peak frequency correlated with an aggregate cognitive change measure (r2= 0.71; p < 0.001), individual cognitive measures, and subjective complaints. Of the combined AEDon/AEDoff patients, 58% exceeded the 95% confidence interval for test-retest change in EEG peak frequency. CONCLUSIONS: Quantitative measures derived from the occipital EEG are sensitive to AEDs and correlate with AED-related cognitive effects and subjective complaints. Although this correlation does not indicate a direct relation, quantified EEG may be a practical measure of AED impact on the brain.