Abstracts

Patients with epilepsy often experience seizures in clusters. We performed preliminary analysis of a prospective study of seizure clustering, to examine risk factors for seizure clustering by history, and the effect of clustering on anxiety, depression and quality of life measures.
Detailed intake questionnaires were administered to all subjects. Seizure clustering was defined as 3 or more seizures in a 24 hr period. Subjects with seizure clustering were subdivided into two groups: Those whose typical seizure pattern was clustering, and those with a history of ever having experienced a seizure cluster. MRI and EEG data were obtained by chart review. Lesional epilepsy was defined as epilepsy associated with significant head trauma, neoplasm, stroke or vascular malformation, excluding MTS. All subjects completed Beck Anxiety Inventories (BAI), Beck Depression Inventories (BDI) and Quality of Life in Epilepsy (QOLIE) testing.
Of the initial 114 subjects recruited, complete data was available for 103 subjects. Overall, 41 subjects (39%) had a history of clustering, including 28 (27%) with a clustering pattern. Lesional epilepsy was significantly associated with clustering (p = 0.019) and with clustering pattern (p = 0.017), however specific MRI findings or EEG abnormalities were not associated with clustering. Poor seizure control was significantly associated with a history of clustering pattern (p = 0.025), but not with overall clustering. In this selected population, there was no significant association between lesional epilepsy and poor control. Mean BAI scores (15.1 cluster group, 15.0 cluster pattern group, 11.7 nonclustered group) and mean BDI scores (17.2 cluster group, 17.3 cluster pattern group, 11.4 nonclustered group) were higher for any history of seizure clustering, but only the mean BDI score in subjects who had ever experienced seizure clustering reached statistical significance (p = 0.015). Similarly, mean QOLIE scores (55.6 cluster group, 62.1 nonclustered group) were lower for any history of seizure clustering, but were of borderline statistical significance (p = 0.086).
Discussion: These preliminary results demonstrate the adverse impact of seizure clustering on prognosis of epilepsy and on quality of life. Those who experienced seizure clustering as the typical seizure pattern are most affected. Further analysis is planned with a larger sample and longer follow up to evaluate the specific predictors of poor and favorable outcome as well as identifying potential areas where intervention may be helpful.
[Supported by: K23 NS02192 (P.I. Dr. Haut)]