Quantitative assessment of neurophysiological signals and motor function following stroke

Abstract

Aim: The aim of this study was to investigate the role of multi-modal neurophysiological signals for assessment and prognosis of patients in sub-acute phases post-stroke. In particular, the relationship between upper limb motor function and quantitative EEG parameters was explored. Methods: Ten male stroke survivors and ten healthy matched controls were included in this study. Bedside clinical examination was done using the Medical Research Council (MRC) scale, its derived motricity index (MI) and the upper limb section of the Fugl-Meyer (FM) assessment score. A multimodal neurophysiological signal acquisition setup consisting of EEG, EMG and muscle forces measurement was used during both upper limb movement and at rest. EEG was further analysed to obtain quantitative parameters including brain symmetry index (BSI) and event related desynchronisation (ERD). Subsequently, two novel parameters were introduced: a long-reference event related desynchronisation (LERD) and an ERD-derived working-hemisphere symmetry index (WHSI). Comparative and correlation analyses were performed between these indices and infarct characteristics, patient motor function and outcome. Six of the stroke survivors were followed-up with repeat sessions monthly. Results: BSI was significantly higher in stroke survivors when compared to healthy subjects, indicating a higher asymmetry following stroke, especially within the cortical stroke subgroup. A statistically significant negative correlation was observed between FM in the follow up session and initial BSI, suggesting possible prognostic value for early BSI. LERD and WHSI demonstrated significant EEG signal attenuation over the stroke-affected hemispheres. These were also inversely correlated with the FM in the follow up sessions. Conclusions: A multimodal neurophysiological signal analysis setup for assessment and prognosis of patients following stroke was designed and successfully realised. Significant differences in conventional (BSI) and novel (LERD and WHSI) EEG parameters were found between stroke and healthy subjects, supporting the validity of the setup used. Above all, the correlations found between FM and initial BSI and LERD parameters suggest a promising role [or these indices to be used as prognostic marker. Keywords: Electroencephalography, stroke, brain symmetry index, event related desynchronisation, Fugl-Meyer assessment score.