Abstracts

Rationale: Our goal was to use behavioral testing to determine if deficits in memory and consciousness have any association with underreporting of seizures; relate these deficits to seizure type and anatomical site of onset. Previous studies suggest that several factors including seizures originating in the dominant hemisphere, impairment in memory and conscious attention may play a role in seizure awareness and reduced patient report of seizures. Little work has been done to relate patients underreporting of their seizures to specific seizure types and anatomical regions.Methods: Patients undergoing seizure evaluation in a VEEG monitoring unit were recruited. Patients responsiveness was assessed by reviewing videos of their seizures. Seizures were classified as generalized tonic-clonic (GTC), complex partial (CPS), or simple partial seizures (SPS) according to conventional definitions. Two reviewers blinded to seizure awareness results filled out a form to assess behavior at onset, during and after the seizures. We compared objective data obtained through VEEG to 3 testing instruments used to verify patients report of their seizures: 1. Admission questionnaire, to assess self-perception of seizures, including questions about seizure types, frequency and awareness 2. Seizure log maintained by patients to record each of their seizures 3. Daily seizure questionnaire administered by research staff to determine awareness of any seizures that occurred in the preceding 24 hours. Clues that may have helped patients realize that they had a seizure were considered (e.g. told by a third party).Results: A total of 93 subjects were recruited to this study. 46 subjects had epileptic seizures and complete questionnaires. There were 23 female and 23 male subjects. Mean age at time of hospitalization was 35 years (range 8-70 years) with 41 adult (aged 18-70) and 5 pediatric (aged 8-17) subjects. Overall we recorded 254 partial seizures comprising 39 secondary GTC, 118 CPS, 43 SPS, and 54 where level of consciousness was not assessed. Patients were unaware of 54% of all recorded seizures including 82% of GTC and 58% of CPS. In contrast, only 14% of all SPS were not identified (p<0.001; chi-square test). Using logistic regression, we found that CPS were significantly less likely to be recognized by patients compared to SPS (OR=0.13, 95% CI=0.04 0.48, p=0.002), and GTC were less likely to be recognized compared to SPS (OR=0.02, 95% CI=0.01 0.10, p<0.001). No significant association was found between preictal sleep vs. wake state and seizure awareness. In addition, we did not observe an effect of side of seizure onset and seizure awareness.Conclusions: This study demonstrates that patients are unaware of more than half of all recorded seizures during inpatient VEEG monitoring. Seizures with impaired consciousness (CPS and GTC) were more likely to go unreported suggesting that consciousness may be a factor influencing the ability of patients to recognize and accurately report their seizures. Further study is needed to investigate the extent to which altered memory function or language impairment contributes to seizure unawareness.