Abstracts

Rationale: Seizure clusters are described as the occurrence of multiple seizures in a short period of time which negatively impacts patient quality of life and often results in increased healthcare utilization. In this study, we performed a preliminary analysis of prospective diary data to determine the prevalence, characteristics and impact of seizure clusters. Methods: Subjects were categorized according to their reported seizure frequency over the year prior to enrollment into one of the three groups: prior clusters, active epilepsy and seizure free. The prior clusters group included subjects reporting a day with two or more seizures. Subjects recorded their particular seizure characteristics—duration, time, frequency, use of rescue medications, occurrence of injuries and emergency room visits—using a paper or electronic diary and were followed on a monthly basis for six months. A patient was considered to have experienced a cluster if there were two or more seizures in a given day or seizures in consecutive days. For patients with daily seizures, a cluster was defined as three times their baseline frequency. Quality of Life in Epilepsy (QOLIE-31) questionnaire was administered at baseline and at six months. Results: Seventy seven out of 231 (33.3%) patients reported seizure clustering in the year prior to enrollment and of those, 19 (24.6%) had been prescribed rescue medication whereas 58 (75.4%) had not. Out of the 231 patients, 60 reached the six month milestone-11 were excluded due to noncompliance with their diaries, providing a sample size of 49. During the six month period, 23 out of 49 patients (47%) experienced seizure clusters. The most prevalent seizure types in clusters were complex partial seizures (17 patients) followed by simple partial seizures (6 patients), generalized tonic-clonic seizures (3 patients) and tonic seizures (2 patients). Five of the 23 patients (21.7%) experienced clusters of two different seizure types. Four out of 49 patients reported the use of rescue medication and all found it to be effective. Seven patients had seizure-related injuries and/or ED visits, of which 4 resulted from cluster events and none used rescue medication. There was no significant difference in the mean total QOLIE-31scores at six months between the prior clusters (63.2, SD 16.3) and active epilepsy groups (58.8, SD 17.4, p=0.69). The seizure free group (84.0, SD 8.92) had a significantly better total QOLIE-31 score compared to the prior clusters (p=0.0001) and active epilepsy groups (p=0.004). Conclusions: Seizure clusters are not uncommon and rescue medications are underutilized. The use of rescue medication may prevent seizure-related injuries and minimize healthcare utilization. Further analysis from this ongoing prospective study may better characterize the prevalence, risk factors and consequences of seizure clusters. (This research was supported by Upsher-Smith Laboratories.)