Abstracts

We have previously shown that carbamazepine (CBZ), oxcarbazepine (OXC), and phenytoin (PHT) can cause slowing of EEG background rhythms and a decrease in objectively measured alertness. EEG slowing was associated with mild negative cognitive test effects. Topiramate (TPM) is a novel AED with known negative cognitive effects. We evaluated the effects of TPM on the EEG and alertness in healthy volunteers. The results were compared with the EEG/cognitive profiles of other AEDs. 40 healthy subjects received 12 weeks of TPM (up to 400mg/day), gabapentin (3600mg/day), or placebo in a double-blind, parallel-group trial. EEG, the Awake Maintenance Task (AMT), and a cognitive battery (including Digit Symbol, Stroop, visual RT, Selective Reminding, Story Recall) were obtained at baseline and at 12 weeks. EEG quantitation was performed using previously described methods. Averaged occipital electrodes were analyzed for peak frequency, median frequency, and percentage theta and delta activity (by power). The AMT was scored for total drowsy time in 6 minutes.
Results were compared with those from healthy volunteers who took CBZ, OXC or PHT in previous 12-week double-blind trials using an identical protocol, and with a group of 70 untreated controls. TPM had no significant effects on EEG background measures, despite having marked effects on cognitive tests ([gt]-2SD median test-retest change for Digit Symbol and Story Recall). GBP slowed the EEG peak frequency (median -0.4 Hz; p[lt]0.01 vs. controls). A Kruskal-Wallis test comparing the EEG effects of all 5 AEDs was significant at p[lt]0.001 (all measures). On 2-group contrasts (Wilcoxon tests) TPM differed from OXC, CBZ, PHT, and GBP (each at p[lt]0.01), each of which significantly slowed the EEG background as compared to untreated controls (p[lt]0.01, peak frequency). TPM had no effect on AMT drowsiness (median test-retest change 0 sec.; NS vs. untreated controls). All other AEDs were associated with a median 20-35 sec. increase in drowsiness (p[lt]0.05 vs. controls).
A plot of EEG vs. cognitive effects for all subjects revealed clusters for TPM, non-TPM AEDs (CBZ, OXC, PHT, GBP), and untreated controls. GBP effects were relatively mild. MANOVA (3 groups) was significant at p[lt]0.001. TPM has a unique profile of neurotoxicity with significant cognitive effects in the absence of any effects on alertness or EEG background rhythms (type 2 neurotoxicity). This profile contrasts with that of CBZ, OXC, PHT, and GBP, each of which are associated with slowing of EEG background rhythms and decreased alertness, in association with mild cognitive effects (type 1 neurotoxicity). (Supported by an independent investigator award from Pfizer Inc.)