Postictal MRI abnormalities and seizure-induced brain injury: Notions to be challenged

Abstract Objective This was a bibliographic search to address the quality of evidence in clinical reports supporting the assertion that brain MRI signal abnormalities are a direct consequence of seizures. Methods The search on PubMed was performed by applying the following inclusion criteria: a) original case reports, b) in humans, c) as single case reports or series of patients, d) of visually detected acute MRI signal abnormalities, e) attributable directly to seizures, and f) published in English. Bibliographic references of initially selected publications were reviewed for additional articles. Full texts of selected publications were read for information regarding clinical, EEG, and MRI features. Moreover, claimed evidence supporting seizure-induced excitotoxicity was assessed. Results The search resulted in 91 publications corresponding to 413 cases. There was a wide range of clinical features and EEG and MRI abnormalities. Premorbid or comorbid conditions were present in many cases, and some of them are potential causes of MRI changes. Claimed evidence for MRI signal abnormalities as a direct consequence of ictal activity was mostly based on the similarity with previous reports, animal models, reversibility, congruent EEG, MRI changes not respecting vascular territories, and ruling out other etiologies. Conclusions Evidence supporting the notion of seizure-induced excitotoxicity is questionable in the studied reports of postictal MRI abnormalities.