Wednesday July 5, 2006 12:00-13:30 Hall 3a Platform Session Neuroimaging

2 C. Barba, 1 D. Di Giuda, 1 D. Policicchio, 1 M. Totaro, and 1 G. Colicchio ( 1 Catholic University, Rome, Italy , 2 Fondazione S. Lucia IRCCS, Rome, Italy ) Purpose: To compare statistical parametric mapping (SPM) with visual analysis of provoked ictal SPECT in defining the epileptogenic zone of drug-resistant epilepsy patients. Method: Twenty patients, submitted to epilepsy presurgical evaluation in the Epilepsy Surgery Unit of Catholic University in Rome, were included in the study. They suffered from drug-resistant epilepsy: 14 symptomatic and 6 cryptogenetic (8 extratemporal, 8 temporal, and 4 multifocal). All patients underwent Technetium-99m Ethyl-Cysteinate-Dimer interictal and pentylenetetrazol provoked ictal SPECT scans. After visual interpretation of interictal and provoked ictal studies, SPM analysis was performed on individual provoked ictal SPECT images in comparison with an established normal brain SPECT database. SPM results were compared to those of qualitative analysis. Results: A good concordance between visual and SPM analysis was found in 19/20 patients with temporal and extratemporal epilepsies. Epileptogenic zone colocalised with the ictal hyperperfusion area in 15 subjects. In 1 case the lateralisation did not correspond, while the other 4 patients were multifocal. Nine subjects were operated with satisfying results (6 Ia, 1 Ib, 1III, and 1 short follow-up). Conclusion: Even though SPM and visual analysis seemed to be comparably sensitive and accurate in localising ictal SPECT foci in both temporal and extratemporal epilepsies, SPM may overcome the inter-observer variability of visual interpretation and allow a more detailed anatomic localisation. The good concordance between qualitative and SPM analysis of ictal studies would suggest the possibility of avoiding performing the interictal SPECT. Finally, provoked ictal SPECT procedure confirmed its efficacy in determining the ictal hyperperfusion area.