Abstracts

Rationale: Periodic lateralized epileptiform discharges (PLEDs) on EEG are most commonly seen in patients with acute cortical lesions but also in those with chronic structural lesions. They are also seen in the setting of focal status epilepticus in the absence of structural lesions on neuroimaging. In this retrospective study we analyze brain MRI changes associated with PLEDs identified on continuous EEG (cEEG) and correlate with clinical outcome. Methods: A retrospective chart review was performed on adult patients admitted between 5/2011 and 5/2012 at a single intensive care unit with PLED pattern seen on cEEG monitoring. Demographics and detailed clinical information (etiology, history of epilepsy, electrographic and clinical seizures, treatment approach and long term outcome) were recorded. Neuroimaging tests were reviewed and findings were categorized as follows: Acute structural lesion (G1); chronic structural lesions (G2); reversible brain MRI changes (seen as DWI hyperintenisty and isointense ADC signal or T2 hyperintensities in the cortical, subcortical and limbic structures that resolved on repeat MRI) (G3); normal brain MRI (G4); and normal CT scan in the absence of brain MRI (G5). Parameters were compared within each radiologically classified subgroup. Results: A total of 31 patients with PLED pattern on cEEG were included in this study. The details of cEEG findings, clinical seizures, history of seizures, interval development of mesial temporal sclerosis on repeat MRI brain and Glasgow outcome scale in different subgroups are described in Table 1. Other characteristic features of the study are described in Table 2. Overall, mortality was 43.7 % in patients with structural lesions (G1+G2), while it was 0% in those with normal or reversible MR changes (G3+G4). Half of the patients with reversible MRI changes had hippocampal atrophy on a repeat MRI scan. Patients with no structural lesions on CT (G5) had the highest mortality at 50%. A high proportion of this subgroup (5 of 8) had co-existing seizures on cEEG. Conclusions: The presence of PLEDs on cEEG is almost always associated with neuroimaging findings on MRI. Patients without structural lesion have better prognosis compared to those with structural lesions. Non-structural and reversible brain MRI can be seen in absence of ongoing seizures on cEEG and may lead to hippocampal atrophy at a later date. This finding suggests ongoing neuronal injury and argues for more aggressive treatment to prevent permanent damage. Larger, prospective studies that use neuroimaging, and more specifically MRI, as part of a decision tree can provide more definitive guidance in treatment approach of PLEDs pattern on cEEG.