Abstracts

Rationale: To describe clinical features and identify predictors of status epilepticus (SE) in hospitalized children admitted with seizures. Methods: The inpatient database for 2008 at our institution was reviewed for patients with epilepsy and seizures. Patients with and without history of SE were compared. We defined SE as seizures lasting ? 5 minutes or recurrent seizures without regaining consciousness between two consecutive seizures for ? 30 minutes. Results: 425 patients (51% male) were identified. Median age of seizure onset was 25 months (0.03-288 months). Median number of semiological seizure types was 2 per patient (1-4). The SE group included 155 patients (36.5% of total; 57% male). These had a previous history of status or new presentation of SE in 2008. Among these, 141 (91%) patients had a history of SE prior to admission and 14 (9%) had no history of seizures presenting with new onset SE. Amongst the 141 patients with history of SE, 80 (56.7%) were admitted in 2008 due to SE, and 61 (43.2%) had a history of SE before 2008. The NON-SE group included 270 patients (63.5% of total; 47.4% male). These did not have SE in the past or during admission. 240 (88.9%) had a history of seizures prior to admission, and 30 presented with new onset seizures on admission in 2008. The median age of seizure onset in both groups was similar (25 months; non-significant). Patients in the SE group presented more often with multiple seizure types (SE 103/155, 66.4%; NON-SE 143/270, 53%; P<0.01). 51 (32.9%) SE patients and 128 (47.4%) NON-SE patients had only one seizure type. 71 (45.8%) SE patients and 96 (35.5%) NON-SE patients experienced two seizure types, and three or more seizure types were identified in 33 (21.2%) SE and 46 (17%) NON-SE patients. SE group patients were prescribed more antiepileptic drugs (AEDs) prior to admission (median 2, range 0-6) than NON-SE patients (median 1, range 0-7; p<0.001). SE group patients also were more likely to have been hospitalized in the past (94; 60.6 %) than NON-SE patients (97; 35.9%; p<0.001). Prior to admission, EEG was available in 396 patients. Spikes were more common in EEGs of SE patients (127/135, 94.1%) than in NON-SE patients (216/261; 82.8%; p<0.01). SE patients also had more frequent slowing on EEG (38/135; 24.5%) than NON-SE (48/261; 18.3%; p<0.05). Amongst 155 SE patients, 134 brain MRIs were available. In the NON-SE group, 244 out of 270 patients had an MRI. Patients with a history of SE more frequently had an abnormal MRI (80; 59.7%) than NON-SE patients (107; 53.7%; p<0.05).