[Seizures in the peracute stage of stroke: incidence and effect on inpatient mortality].

INTRODUCTION

According to literature data, the effect of seizures on inpatient mortality has not yet been not definitly determined and data on seizure incidence significantly vary with different methods used in various studies,

AIM

Our aim was to determine the effect of seizures in the peracute stage of stroke on inpatient mortality and to detect the possible association of seizure incidence and stroke subtype, age and sex.

SUBJECTS AND METHODS

Medical documentation of patients hospitalized at the ward for cerebrovascular diseases, Department of Neurology, Sveti Duh General Hospital from 1998 to 2005 was retrospectively analyzed. There were 3,542 patients with hemorrhagic or ischemic stroke. Seizures were defined as seizures occurring immediately before or within the first 24 hours of the neurological deficit development. Patients were subdivided into two groups, with and without seizures. Patients with a previous history of seizures were excluded from the study.

RESULTS

In a total of 3542 patients, there were 1885 (53.21%) male and 1657 (46.78%) female patients, mean age 70.06+11.03 in the non-seizure group and 67.93 +/- 13.02 in the seizure group. Seizures were significantly more frequent (p = 0.0002, OR 2.63) in patients with hemorrhagic stroke. Seizures were observed in 42 of 2931 (1.43%) patients with ischemic stroke and in 23 of 611 (3.76%) patients with hemorrhagic stroke. Total inpatient mortality was 21.4% in the group without seizures, and 30.8% in the group with seizures (p = 0.04). The ischemic stroke inpatient mortality in the seizure group was 19.5% (17.4% in non-seizure group), and hemorrhagic stroke inpatient mortality was 52.2% in the seizure group (41% in non-seizure group). There was no significant age and sex difference between the groups with and without seizures.

DISCUSSION

The rate of seizures has been estimated 1.7% to 8.9%. Different incidence rates are the result of different methods and different seizure definitions used in various studies. In our study seizures were significantly more frequent in patients with hemorrhagic stroke than in those with ischemic stroke, which is comparable to other studies. The slightly higher percentage of seizures reported in other studies is the result of the fact that they have analyzed seizures occurring from 48 hours to up to 30 days from stroke onset. The influence of seizures on inpatient mortality remains uncertain since different studies used different definitions of seizures and analyzed different stroke subtypes. Some studies found that seizures were associated with increased inpatient mortality; after accounting for stroke severity, population-based studies found no association between seizures and lethality. Our results support the conclusion that seizures cannot be taken as a prognostic factor in stroke.

CONCLUSION

The incidence of seizures during the first 24 hours of stroke onset was statistically significantly higher in patients with hemorrhagic stroke than in those with ischemic stroke. When all patients irrespective of stroke subtypes were analyzed, total inpatient mortality was statistically significantly higher in the seizure group. However, there was no statistically significant difference in inpatient mortality when ischemic and hemorrhagic groups were analyzed separately. There was no age and sex difference in the rate of seizures,