Outcomes and responses to anti-seizure medications in status epilepticus: The significance of peri-ictal EEG and MRI abnormalities

Abstract

Background

Peri-ictal MRI (PMA) and EEG abnormalities (PEA) may predict the outcomes and responses to anti-seizure medications (ASMs) in patients with status epilepticus (SE).

Methods

The inclusion criteria were as follows: (1) patients with SE aged >18 years, (2) peri‑ictal MRI including DWI and FLAIR, (3) peri‑ictal EEG, (4) well-documented medical records, and (5) recruitment between January 2016 and December 2023. Patients with hypoxic-ischemic encephalopathy as the etiology were excluded. Outcomes were classified as good (modified Rankin Scale 0–3) or poor (modified Rankin Scale 4–6) at discharge. Response to ASMs was defined as a failure if SE was not controlled with second-line ASMs at standard doses. PEA was defined as electro-clinical seizures, ictal-interictal continuum, or lateralized periodic discharges. PMA was defined as increased signal intensities in the cortex, thalamus, or hippocampus. A model was developed to predict poor outcomes using multivariate logistic regression with PMA, PEA, and response to second-line ASMs as dependent variables.

Results

A total of 114 patients were included. Poor outcomes were observed in 43.9 % (50/114), and 31 % (35/114) were refractory to second-line ASMs. A multivariate logistic regression predicting poor outcomes identified four variables that were independently associated with poor outcomes (p < 0.0001, R²=0.46): PEA (OR 4.4, 95 % CI 1.68–11.31), PMA (OR 3.8, 95 % CI 1.22–12.1), failure of second-line ASMs (OR 3.7, 95 % CI 1.29–10.84) and status epilepticus severity score (OR 1.6, 95 % CI 1.16–2.13).

Conclusions

PMA, PEA and failure of second-line ASMs were robust predictors of poor outcomes in patients with SE.