An epileptic seizure is a frequent complication of Stroke - the most common etiology of epilepsy in patients over 60 years15. Acute symptomatic poststroke seizures are believed to result from cellular biochemical dysfunction by ischemia leading to increased extracellular concentrations of glutamate, a neurotransmitter associated with neuronal injury, generating electrically irritable tissue16–18.
In our sample, the prevalence of epileptic seizures was similar to that observed in prospective studies and systematic reviews of acute seizures after stroke6,10,11,19−22. Most of these studies considered crisis occurring up to 7 to 15 days after the vascular event as acute6,10,11,19. A lack of systematic evaluation of our patients with continuous electroencephalography can explain a low prevalence in comparison to studies with large samples18.
We did not find gender and age differences in patients with and without seizures. These data are consistent with previous studies17,20; although men and youth have shown to be risk factors in some studies5,17,20,23. This result may be explained by some baseline characteristics of younger patients such as the volume of ischemia and more recognizable epileptic features18,22. We did not find divergences in clinical baseline aspects between elders and youngers.
Similarly, to some authors11, in our work, we found a 1.6x higher prevalence of TACS in this group; strengthening the widely reported relationship between stroke with cortical large lesions and seizure7,8,10,11,16−24.
Following the literature, we also found in our sample that a higher NIH Stroke Scale score on admission, greater the risk of acute symptomatic crisis. Although different risk factors have already been mentioned in the literature, stroke severity, translated by the NIH Stroke Scale score, is associated with a higher risk of seizures in several studies10,11,19,22.
In our research, we demonstrated a higher risk of crisis associated with the presence of symptomatic HT. We suppose that symptomatic HT is strongly associated with large and cortical ischemic areas and some surrounding conditions such as aging, cardioembolism and Diabetes Mellitus – these conditions may be associated with epileptic seizures regardless of HT5,10,16,19. Moreover, the presence of blood in cerebral ischemia may deflagrate an inflammatory cascade leading to brain edema and impaired neuronal homeostasis5.
We did not find a higher prevalence of cardioembolic stroke in patients with epileptic seizures, agreeing with some studies that do not bring this association. On the other hand, some authors have reported this association24 − 16 – we suppose that high collinearity between cardioembolism and hemorrhagic transformation in our sample could lead to a lack of association.
The relationship between epileptic seizures after stroke and the worse outcome is still uncertain10–12. Numerous clinical reports pointed out the increased length of hospital stay and higher morbidity and mortality among patients who develop seizures.5,9,13. We found that the average length of stay of patients with the crisis was twice those without crisis. However, we found no relationship between crisis and unfavorable outcomes, including death, as in some studies17,19.
Our study has many limitations. First, we performed a retrospective analysis from our local dataset, which included only patients not submitted to reperfusion therapies (RT). We suppose that differences in baseline aspects between patients who were submitted or not to RT would lead to another result from what we have found. Second, we did not obtain the exact time between the vascular event and the onset of seizures; therefore, we need to consider epileptic seizures as acute given the patients' average length of stay of 18 days and the definition of acute symptomatic seizure as in any period during the hospitalization for an ischemic stroke. Third, we classified those with epileptic seizures based on clinical manifestations corroborated by electroencephalographic analysis – a continuous electroencephalographic analysis was not adopted in those without clinical manifestations. Finally, we analyzed preselected clinical complications retrospectively and it was not possible to document the effect of the others (i.e. pain, pulmonary embolism, hyperglycemia).