Long-term follow-up of post-stroke epilepsy after ischemic stroke: Room for improved epilepsy treatment

Petra Redfors; Lukas Holmegaard; Annie Pedersen; Christina Jern; Kristina Malmgren

doi:10.1016/j.seizure.2020.01.009

Long-term follow-up of post-stroke epilepsy after ischemic stroke: Room for improved epilepsy treatment

Seizure. 2020 Jan 21:76:50-55. doi: 10.1016/j.seizure.2020.01.009. Online ahead of print.

Authors

Petra Redfors¹, Lukas Holmegaard², Annie Pedersen³, Christina Jern³, Kristina Malmgren²

Affiliations

¹ Institute of Neuroscience and Physiology, Sweden. Electronic address: petra.redfors@vgregion.se.
² Institute of Neuroscience and Physiology, Sweden.
³ Institute of Biomedicine, The Sahlgrenska Academy at University of Gothenburg, Sweden.

PMID: 32014726
DOI: 10.1016/j.seizure.2020.01.009

Abstract

Purpose: To assess long-term incidence and predictors of post-stroke epilepsy (PSE) and to evaluate the antiepileptic drug (AED) treatment in a well characterized cohort of middle-aged patients.

Methods: The study is based on the Sahlgrenska Study on Ischemic stroke, and included 1066 adult patients with first-ever or recurrent acute ischemic stroke (AIS) before the age of 70. Early seizures (ES) were defined as seizures within one week and PSE as unprovoked seizures occurring more than one week from index stroke. Cardiovascular risk factors, subtype of AIS, and stroke severity were determined at baseline. ES, PSE, treatment with AEDs, recurrent stroke and mortality were assessed through national registers and medical records. Cox regression models were used for identifying predictors of PSE.

Results: Twenty-six patients (2.4 %) developed ES. After a median follow-up of 8.0 (IQR 4.1-10.9) years, 84 (7.9 %) had PSE, and 160 (15.0 %) had experienced a non-fatal recurrent stroke. Stroke location (total anterior and partial anterior circulation infarct, both P < 0.001), ES (P < 0.001), stroke recurrence (P < 0.001), artery dissection (P < 0.002), and previous coronary heart disease (P < 0.006) were independent predictors of PSE. Only 10 (11.9 %) had the first seizure more than four years after index stroke. In 24 (30 %) PSE patients, seizure control was not achieved.

Conclusions: In addition to well-known risk factors for PSE development, our data also identified stroke recurrence, artery dissection and established coronary disease. Seizure control was less common than expected and in a significant proportion of patients AEDs had not been adjusted despite continuing seizures.

Keywords: Anticonvulsants; Cerebral infarction; Follow-up studies; Incidence; Seizures.