EEG signal varies with different outcomes in comatose patients: A quantitative method of electroencephalography
Weibi Chen1, Gang Liu1, Yingying Su1
1Department of Neurology, Xuanwu Hospital, Capital Medical University, Beijing, China.
Background:
Electroencephalographic reactivity (EEG-R) is a major predictor of outcome in comatose patients; however, the inter-rater reliability is limited due to the lack of homogeneous stimuli and quantitative interpretation.
New Methods:
EEG-R testing was employed in comatose patients by quantifiable electrical stimulation. Reactivity at different frequency bands was computed as the difference between pre- and post-stimulations in power spectra and connectivity function (including magnitude squared coherence and transfer entropy). The clinical outcomes were dichotomized as good and poor according to the recovery of consciousness. Signal discrimination of EEG-R was compared between the two groups.
Results:
A total of 18 patients (43%) regained consciousness at a 3-month follow-up. In the patients who regained consciousness, the EEG power increased significantly (P < 0.05) at the Alpha and Beta frequency bands after stimulation as compared to those with no behavioral awakening. Also, connectivity enhancement (including linear and nonlinear analysis) in the Beta and Gamma bands and connectivity decrease (nonlinear transfer entropy analysis) in the Delta band after stimulus were observed in the good outcome group.
Comparison With Existing Method(S):
In this study, the combined use of quantifiable stimulation and quantitative analysis shed new light on differentiating brain responses in comatose patients with good and poor outcomes as well as exploring the nature of EEG changes concerning the recovery of consciousness.
Conclusions:
The combination of quantifiable electrical stimulation and quantitative analysis with spectral power and connectivity for the EEG-R may be a promising method to predict the outcome of comatose patients.
