Abstracts

Rationale: The event-related potentials (ERP) technique defines covert cognitive processes and quantifies neural processing speed with exquisite temporal sensitivity. We hypothesized N2pc, an ERP measure of visual attention, would be delayed in epilepsy, and sought to analyze the relationship between specific independent clinical predictor variables (age, sex, epilepsy syndrome, antiepileptic drug load) and the dependent variable of visual attention processing time in epilepsy patients. Methods: N2pc ERP responses were recorded from 30 controls and patients (10 mesial temporal lobe, 10 extratemporal partial, and 10 primary generalized epilepsy patients receiving 0-4 antiepileptic drugs (mean=1.57)). Subjects detected targets in right or left visual fields during a feature-conjunction visual-search task. Analysis compared ERP waves from equivalent EEG channels ipsilateral and contralateral to the target. Fractional area and fractional local peak latencies (FAL, FLPL) were then determined. Group comparisons were analyzed by 1-way ANOVA, and clinical predictor variables (age, sex, epilepsy type, and antiepileptic drug load) were regressed onto FAL utilizing multivariable linear regression analysis in SPSS (Chicago, IL). Results: Epilepsy patients showed significantly delayed N2pc FAL and FLPL compared to controls. FAL in channels O12/OLR was delayed by 30.8 ms (F=8.98, p=.006), and N2pc FLPL in channels O12/OLR was delayed by 38.9 ms (F=6.00, p=.021). Only antiepileptic drug load was significantly related to N2pc FAL (F=10.226, p=0.004). Conclusions: Visual attention was significantly delayed in epilepsy by a degree that could cause substantial problems under natural viewing conditions. Antiepileptic drug load predicted visual attention impairments in epilepsy. Future studies in epilepsy patients will further define which antiepileptic drugs and doses are most predictive of impaired visual attention.