Abstracts

Rationale: Zonisamide (ZNS) is a new antiepileptic drug with several mechanisms of action, including voltage sensitive sodium channels, blockade of T-type calcium channels and also an allosteric modulation of GABA(A) receptor. The Japanese, US and European data provide a large database of safety information suggesting that zonisamide is safe, effective and generally well tolerated as monotherapy and adjunctive therapy in epileptic patients with various seizures types. However, some authors have described cognitive adverse effects, such as language and verbal impairment in epileptic patients treated with zonisamide, similar to the ones described with drugs of similar chemical structrure such as topiramate. Methods: We report preliminary findings in language fMRI and neuropsychological evaluation in a group of eight refractory epileptic patients before and after add-on treatment with ZNS. Language neuropsychological evaluation and functional neuroimaging were performed in twelve refractory epileptic patients before and after three months of add-on treatment with zonisamide at doses > or =300mg, without modification of concomitant antiepileptic drugs. Results: Because of technical difficulties and inadequate cooperation during language tasks in some patients, we could analyze the results of eight patients only. Wilcoxon matched pair test analysis of the obtained data from neuropsychological test and fMRI before and after the treatment, was performed to evaluate the impairment or improvement in the scores of neuropsychological tests and number of significant voxel activation in fMRI (p<0.001 uncorrected and cut-off p<0.05 corrected for multiple comparisons) over bihemispheric language regions of interest. We found a trend towards a decrease (p<0.06) in the number of voxel activation during word generation tasks in contrast to a significant increase in number of voxel activation (p<0.05) in semantic-decision tasks in fMRI. On the other hand, a significant impairment (p<0.03) in word generation neuropsychological test was observed, as well as, a significant improvement (p<0.02) in neuropsychological tests involving semantic-decision. Conclusions: Add-on treatment with ZNS may have complex effects over language functions in patients with epilepsy. As evidenced by neuropsychological testing and language fMRI, ZNS may impair word generation tasks but improve semantic language. This may explain the overall good cognitive profile of the drug. Functional magnetic resonance imaging may help to assess the dynamics of cognitive changes associated to the use of AEDs.