Cortical connectivity from EEG data in acute stroke: A study via graph theory as a potential biomarker for functional recovery

Abstract

Cerebral post-stroke plasticity has been repeatedly investigated via functional neuroimaging techniques mainly based on blood flow/metabolism. However, little is known on predictive value of topological properties of widely distributed neural networks immediately following stroke on rehabilitation outcome and post-stroke recovery measured by early functional outcome. The utility of EEG network parameters (i.e. small world organization) analysis as a potential rough and simple biomarker for stroke outcome has been little explored and needs more validation. A total of 139 consecutive patients within a post-stroke acute stage underwent EEG recording. A group of 110 age paired healthy subjects constituted the control group. All patients were clinically evaluated with 3 scales for stroke: NIHSS, Barthel and ARAT. As a first result, NIHSS, Barthel and ARAT correlated with Small World index as provided by the proportional increment/decrement of low (delta) and viceversa of high (beta2 and gamma) EEG frequency bands. Furthermore, in line with the aim of the present study, we found a strong correlation between NIHSS at follow up and gamma Small World index in the acute post-stroke period, giving SW index a significant weight of recovery prediction. This study aimed to investigate possible correlations between functional abnormalities of brain networks, measured by small world characteristics detected in resting state EEG source investigation, and early post-stroke clinical outcome in order to find a possible predictive index of functional recovery to address and/or correct the rehabilitation program.