Abstracts

Rationale: Patients with temporal lobe epilepsy (TLE) often demonstrate memory dysfunction in the setting of otherwise normal intelligence. This finding is consistent with the theory that establishing new memories is dependent to a large extent on the function of the hippocampus and other mesial temporal lobe structures, a network known to be dysfunctional in TLE. Prior studies found an association between disruptions of cognitive task performance and the presence of interictal epileptiform discharges (IEDs). Little is known, however, regarding the properties of IEDs that pose the greatest risk to cognitive function. The present study examines the effects of IEDs during encoding on subsequent recall and explores the roles of laterality, spatial extent, and duration. Methods: Subjects with suspected TLE completed an incidental memory task during intracranial EEG recordings with hippocampal depth electrodes placed for clinical purposes. Subjects viewed a series of images on a computer monitor and were asked to make a decision about each item (i.e. whether it was used indoors or outdoors). After a brief distracting task, in which subjects classified a series of faces as male or female, participants recalled as many items from the encoding list as possible (Figure 1). IEDs were identified by manual review. The percentage of items later recalled in the presence or absence of IEDs during encoding was compared using a Wilcoxon signed-rank test. Data were also stratified by lateralization. To determine the effects of spatial extent of IEDs and discharge duration on recall performance, the number of electrodes demonstrating IEDs and the amount of time encompassed by discharges during each encoding trial were calculated. The relationships between the number of involved electrodes, discharge duration, and memory performance were assessed using logistic regression. Results: Six subjects participated in the study (2 male, 4 female, aged 22-57 years, 5 right handed, 1 left handed). Four subjects received right hippocampal depth electrodes, one subject received left hippocampal depth electrodes, and one subject was implanted bilaterally. Four subjects had seizure onset zones that included the temporal lobe, while two subjects did not have spontaneous seizures recorded during the clinical evaluation. Median percent correct recall was 17.2% (range 10.4-21.9%). Median percent correct recall in the presence or absence IEDs is shown in Figure 2, for all IEDs combined, as well as for left and right IEDs considered separately. No statistically significant effects of presence vs. absence of IEDs, spatial extent, or duration were observed (all p values >0.05). Conclusions: The effect of the presence vs. absence of IEDs was small and not statistically significant. Nor did the spatial extent or duration of IEDs predict recall performance. These findings may be due to floor effects of overall poor task performance, limited variability in discharge characteristics, and a brief duration of discharges relative to stimulus presentation. There was, however, a pattern suggestive of poorer performance in the setting of IEDs, particularly with left sided discharges. This finding may reflect the verbal memory component of this free recall task. A larger sample size would be necessary to determine whether this pattern is consistent. Funding: This work was supported by Career Development Award number IK2 CX-001255-01 from the United States (U.S.) Department of Veterans Affairs Clinical Sciences R&D (CSRD) Service (BAL).