Network electroencephalographic (EEG) measures in primary progressive apraxia of speech and aphasia: Biomarkers (non‐neuroimaging) / Differential diagnosis. (7th December 2020)

• Record Type: Journal Article
• Authors: Utianski, Rene L; Botha, Hugo; Caviness, John N; Worrell, Gregory A; Duffy, Joseph R; Clark, Heather M; Whitwell, Jennifer L; Josephs, Keith A
• Is Part Of: Alzheimer's & dementia. Volume 16(2020)Supplement 5
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-12-07
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.039056
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Past research has demonstrated that clinical EEG is sensitive to dementia associated with Alzheimer's (AD) and Parkinson's diseases (PD), and nonfluent/agrammatic Primary Progressive Aphasia (agPPA), but not Primary Progressive Apraxia of Speech (PPAOS). However, clinical EEG studies do not quantify interactions of multiple brain areas, or network activity. Graph theory (GT) simplifies analyzing and comparing complex brain networks. Studies have shown changes in EEG GT measures in AD and PD, but have not been studied in agPPA or PPAOS. The goal of this study was to use GT to characterize network‐level changes and associations with speech and language functioning in patients with agPPA and PPAOS. Methods: Scalp EEGs of 14 agPPA and 7 PPAOS patients were collected during relaxed wakefulness (10‐20 positions, 256Hz sampling, 1–200Hz bandpass filter). Eight artifact‐free epochs were selected from non‐overlapping 1024‐point epochs. Via Brainwave software, GT weighted connectivity and minimum spanning tree (MST) measures were calculated for theta and alpha frequency bands. Differences in theta MST measures between agPPA and PPAOS were assessed with Wilcoxon rank‐sum tests. Spearman correlations were computed between behavioral and all network measures across all patients. Results: Theta frequency MST diameter and eccentricity were significantly lower and leaf was significantly higher in agPPA compared to PPAOS patients; betweenness centrality did not differ. This indicates agPPA networks are more integrated (star‐like) than PPAOS. There were significant correlations between several network measures (e.g. kappa, lambda, PLI) and indices of speech‐language functioning. Conclusions: The results provide EEG evidence of network alteration in agPPA and PPAOS patients. The GT measures and associations with behavioral measures provide support for reduced global efficiency, or network specificity, in these patients. The correlation of language measures with lambda (path length) in the theta frequency band suggests agPPA is associated with an inefficient path length and/or increased integration. The correlation between apraxia of speech (AOS) measures and MST eccentricity in the alpha frequency band suggests that more severe AOS is associated with reduced distance between two nodes. This study demonstrates potential for EEG GT measures to quantify network changes and distinguish network changes between related but different clinical syndromes.