Abstracts

Rationale: Often routine clinical EEG and MRI do not adequately define the ictal focus in surgical planning of focal epilepsies. SPECT and high resolution (dense array) EEG are two strategies that allow better imaging of seizure foci. We have used a strategy of combining both types of studies in a number of patients undergoing pre-surgical evaluation. We compare the clinical yield of these two modalities here.Methods: Fifteen patients had both dense array EEG (DA-EEG) recording and an ictal SPECT injection of their seizures (often during the DA-EEG monitoring). Additional patients who did not have a seizure for SPECT injection were excluded. SPECT and DA-EEG interpretations were performed by separate investigators often blinded to the other result. Since the accuracy of the ictal localizations have not been invasively verified, only lateralized changes are compared here. Results: Nine seizures on average (range 1 to 20) were recorded with DA. Each patient only had one 'successful' SPECT injection. Many of the seizures were captured at times SPECT injection would not have been possible (e.g. late in the evening). Results are presented in the accompanying table. Hemispherically lateralized SPECT changes were seen in 13 patients. With DA-EEG, 11 patients had lateralized EEG changes. In one patient DA-EEG could not lateralize changes occurring near midline parietal region, although the vicinity was comparable to SPECT; SPECT resolved the laterality. SPECT was not lateralized in one subject in whom DA was lateralized; this patient later had successful surgical treatment corresponding to DA-EEG data. In three subjects ictal SPECT was lateralized, but the DA EEG was not; the accuracy of SPECT lateralization in these cases has not been verified yet. In two patients the lateralization between SPECT and DA-EEG were contradictory. In one of the cases semiology was in agreement with DA-EEG and was confirmed by intracranial monitoring; unfortunately the patient did not become seizure free. In the other case invasive monitoring could not lateralize the electrographic changes either and surgery was not possible. Conclusions: Dense Array EEG and SPECT often show similarly lateralized EEG changes when previous testing has failed. In cases of disagreement, the invasive monitoring results were also low yield, indicating the complexity of the cases involved. Outcome data regarding the comparison of the accuracy of the localization by these two methods is still pending. DA-EEG offered the additional advantage of capturing multiple seizures occurring at any time, whereas SPECT is not avaiable at night in most institutions. Overall, these results show the two modalities are in often in agreement and present complementary clinical information.