Introduction: Cerebrovascular diseases are the most frequent risk factor for epilepsy in the elderly, and epileptic phenomenon following stroke is known to worsen the prognosis. Although electroencephalography is the gold standard epilepsy biomarker, it is rarely used in post-stroke studies, and the frequency of post-stroke epileptiform activity is still uncertain.
Patients and methods: We analysed studies indexed to MEDLINE, Embase, Web of Science, PsycINFO and OpenGrey (up to March 2015), reporting post-stroke electroencephalographic epileptiform activity frequency in adults. Epileptiform activity was classified as ictal (electrographic seizures) and interictal (non-periodic spikes and sharp waves). Data selection, extraction and appraisal were done in duplicate. Random-effects meta-analysis was used to pool frequencies.
Results: The pooled frequency of post-stroke ictal and interictal epileptiform activity was 7% (95% CI 3%-12%) and 8% (95% CI 4%-13%), respectively. The use of continuous electroencephalogram was not associated with an increased frequency of electrographic seizures (p = 0.05), nor did the management setting (Intensive Care Unit versus non- Intensive Care Unit, p = 0.31). However, studies with continuous electroencephalogram showed a higher frequency of interictal epileptiform activity (p = 0.01).
Discussion: This study provides the best available estimates of the frequency of post-stroke electroencephalographic epileptiform activity. Due to detection bias, it was not possible to correlate clinical and electrographic seizures.
Conclusion: The frequency of ictal and interictal epileptiform activity in the electroencephalogram was comparable with previous frequency analyses of clinical seizures. The frequency of ictal epileptiform activity did not change with continuous record or clinical setting, while the frequency of interictal epileptiform activity increased with continuous record.
Keywords: Epileptic seizures; electroencephalogram; meta-analysis; stroke; systematic review.