Intracortical EEG for the detection of vasospasm in patients with poor-grade subarachnoid hemorrhage

Abstract

Background: To study the feasibility of utilizing intracortical electroencephalography (ICE) including quantitative EEG (qEEG) analysis for the detection of vasospasm in five consecutive poor-grade SAH patients.

Methods: Intracortical electroencephalography (ICE) was obtained via a single miniature parenchymal 8-contact depth electrode placed at the bedside. Quantitative EEG parameters, calculated on surface EEG and ICE, included alpha/delta ratio (ADR), mean amplitude, suppression percent, and total power. Percent changes between averaged values over 4-6 h of baseline EEG and EEG prior to angiography were calculated. The entire continuous qEEG recording for each patient was then reviewed to determine optimal automated alarm criteria.

Results: ICE ADR was the most accurate for predicting angiographic vasospasm (5/5). ICE ADR decreased between baseline and follow-up by 42% (from 0.56 ± 0.07 to 0.32 ± 0.03) for those with vasospasm (N = 3) compared to 17% (0.62 ± 0.06 to 0.51 ± 0.03) for those without (N = 2). A sustained decrease in the ICE ADR from baseline (>25% for ≥ 4 h) occurred in all three patients with angiographically confirmed vasospasm and not in the two without; this decline occurred 1-3 days prior to angiographic confirmation.

Conclusions: Intracortical EEG is promising for detecting ischemia from vasospasm in poor-grade SAH patients, may be superior to scalp EEG, and allow automated detection, particularly using the ADR. Larger studies are needed to better define the effectiveness of this approach.