Related Experiment Video
Updated: May 12, 2026
Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms
Published on: November 1, 2019
RSE prediction by EEG patterns in adult GCSE patients
Fei Tian1, Yingying Su, Weibi Chen
1Department of Neurology, Xuanwu Hospital, Capital Medical University, No 45 Changchun Street, Beijing 100053, China.
Background:
Electroencephalogram (EEG) can predict mortality in status epilepticus (SE) patients. However, we consider that the prediction for refractory status epilepticus (RSE) after SE initial treatment is more significant than long-term prognosis of SE. The objective of this study is to detect some predictive EEG patterns for RSE.
Methods:
Pooled data derived from two randomized controlled trials (RCTs) were prospectively analyzed in adult generalized convulsive status epilepticus (GCSE) patients.
Results:
Etiology, GCSE duration and EEG patterns are three factors which were statistically different between non-RSE and RSE groups. However, when we introduced these factors into multivariable logistic regression model, only EEG pattern was an independent risk factor for RSE prediction. Comparing with rhythmic fast activities background (RFAB) pattern, there were positive correlations between interictal epileptiform discharges (IEDs), periodic epileptic discharges/subtle status epilepticus (PEDs/subtle SE) patterns and RSE incidence respectively.
Conclusions:
There was an increased risk of RSE incidence accompanied with IEDs and PEDs/subtle SE patterns appearance. Clinicians should adjust anti-epileptic strategies with the aid of these EEG patterns in order to reduce RSE incidence.
