Abstracts

Rationale: It is well-known that stroke is an important cause of epileptic seizures. The literature is scarce when it tries to assess the influence of seizures after and at stroke onset. The aim of the study was to evaluate the impact of seizure occurrence in the outcome after ischemic strokes, as well as to determine predictors of seizure in those patients who have strokes.Methods: This was a cohort study that included consecutive patients with acute stroke between July 2003 and March 2008 from the Registry of the Canadian Stroke Network (RCSN), the largest database of patients in Canada with acute stroke seen in 11 selected acute care hospitals. We compared clinical characteristics and outcomes in those patients with seizures and those without. In the group of patients with seizures we analyzed the same variables among those with seizures at onset and those with seizures during hospitalization, after onset of stroke. We identify predictors of seizures, as well as measures like mortality, length of stay, discharge disposition, and use of health-related resources. Chi square was used to compare categorical variables while t-test and median two-sample test were used for continuous variables. Logistic regression analyses, including step-wise methods were used to assess the relationship seizures-stroke, and to identify predictors of seizure occurrence. We used SAS. All tests were two-tailed and p-values <0.05 were considered significant.Results: From 25,671 patients included in the RCSN, we included 10,261 patients (excluded: non-admitted patients, non-ischemic strokes, non-first stroke event, and invalid health card numbers). Mean age was 72.2 (SD +/- 0.1), 48.2% were females, and most were Non-Hispanic Caucasians (56.1%). 157 patients (1.5%) had a seizure at stroke presentation, while 208 had a seizure after it (2%). Diabetes, alcohol consumption, and Charlson index were not different among the two groups and those without seizures. Depression (p=0.02), neglect, pneumonia, and UTI (p<0.01) were seen more frequently in those with seizures after stroke onset (p=0.02). Motor deficit was most commonly seen in those with seizure at stroke presentation (p<0.01). Higher use of health-related resources was seen in those with seizures at stroke presentation (p<0.01). In general, seizures were seen more common in those with more severe strokes (p<0.01), and mortality and disability was worse in them (p<0.01). Predictors of seizure at onset included: age (>