Functional Brain Network Topology Discriminates between Patients with Minimally Conscious State and Unresponsive Wakefulness Syndrome

Alberto Cacciola, Antonino Naro, Demetrio Milardi, Alessia Bramanti, Leonardo Malatacca, Maurizio Spitaleri, Antonino Leo, Alessandro Muscoloni, Carlo Vittorio Cannistraci, Placido Bramanti, Rocco Salvatore Calabrò, Giuseppe AnastasiJournal of Clinical Medicine 2019 · DOI

TL;DR

This study uses EEG-based functional brain network analysis to identify distinct topological patterns between patients in a Minimally Conscious State (MCS) and those in an Unresponsive Wakefulness Syndrome (UWS). The findings reveal altered interhemispheric fronto-parietal connectivity and specific network topology features in UWS patients, suggesting these network characteristics could serve as objective markers for differentiating these states of consciousness.

What Problem Does This Paper Address?

Consciousness is a complex emergent property of the brain, arising from the integrated functioning of multiple neural structures. Disorders of Consciousness (DoC), such as UWS and MCS, represent a significant challenge in clinical diagnosis and prognosis. While previous research has implicated fronto-parietal networks in consciousness, it remains unclear how the overall organization and topological features of these brain networks differ between these states. Accurately distinguishing between UWS (where patients show no signs of awareness) and MCS (where patients exhibit inconsistent but discernible signs of awareness) is crucial for appropriate patient care and rehabilitation. Current diagnostic methods, like the Coma Recovery Scale-Revised (CRS-R), can be subjective and prone to misdiagnosis. This paper aims to address this by exploring whether network topology measures derived from resting-state EEG can provide objective, group-level distinctions between UWS and MCS patients.

Methods

The researchers employed a sophisticated network science approach to analyze resting-state EEG data. Key methods include:

  • EEG Recording and Preprocessing: Standard 19-channel EEG was recorded from patients at rest. Data underwent rigorous artifact removal and filtering (0.7-30 Hz).
  • Source Localization and Functional Connectivity: eLORETA software was used to estimate electrical neuronal activity across 84 regions of interest (ROIs). Functional connectivity was then quantified using Peak Lagged Phase Synchronization (LPS) across different EEG frequency bands.
  • Network Construction: LPS values were used to construct weighted functional connectivity matrices for each patient. These matrices were then thresholded using proportional thresholding (PT%) to create binary graphs, focusing on the strongest connections. A specific threshold of 20% was chosen for detailed analysis.
  • Network Topology Analysis: A comprehensive suite of network measures was applied, including:
  • Deterministic Measures: Characteristic path length, average clustering coefficient, global efficiency, local efficiency, node betweenness centrality, and local-community-paradigm correlation (LCP-corr).
  • Stochastic Measures: Small-worldness (SWw and SWw-E).
  • Network-Based Statistics (NBS): This approach was used to identify specific subnetworks with significant differences in functional connectivity between UWS and MCS groups.
  • Statistical Analysis: Mann-Whitney U tests were used to compare network measures between groups. Receiver Operating Characteristic (ROC) curves were employed to assess the discriminative power of these measures. Correlations with clinical scores (CRS-R) were also examined.

Datasets

The study utilized resting-state EEG data from 45 patients diagnosed with disorders of consciousness (DoC) attending the IRCCS Centro Neurolesi Bonino Pulejo in Messina, Italy. These patients were categorized into two groups:

  • Minimally Conscious State (MCS): n = 13
  • Unresponsive Wakefulness Syndrome (UWS): n = 12 (Note: 6 patients were excluded due to poor EEG data quality, resulting in 25 analyzed EEG datasets).

The study did not introduce a new benchmark dataset but rather analyzed existing patient data to investigate established diagnostic categories.

Key Insights and Ideas

  • Interhemispheric Fronto-Parietal Network as a Key Differentiator: The study strongly suggests that the integrity of the interhemispheric fronto-parietal network, particularly in the beta-1 (β1) frequency band, is crucial for distinguishing between UWS and MCS. Decreased connectivity in this network was a hallmark of UWS patients.
  • Aberrant Local Organization in UWS: UWS patients exhibited altered local network topology, characterized by increased local-community-paradigm correlation (LCP-corr), higher clustering coefficient, and increased local efficiency. This suggests a shift towards more segregated and locally focused processing in UWS, potentially indicating a compensatory mechanism or a disruption in global integration.
  • Nodal Level Alterations: Specific brain regions within the limbic and temporo-parieto-occipital areas showed altered functional topology (degree, betweenness centrality, clustering coefficient) in UWS patients, highlighting localized disruptions in information processing.
  • EEG as a Viable Tool: The study demonstrates that resting-state EEG, combined with advanced network analysis, can provide valuable insights into the functional connectome of DoC patients, offering a potentially more accessible clinical tool compared to fMRI.

Key Results

  • Network-Based Statistics (NBS): Identified a significant subnetwork with decreased functional connectivity in UWS patients compared to MCS patients in the β1 frequency band (p = 0.004). This subnetwork primarily involved interhemispheric fronto-parietal connections.
  • Whole-Brain Network Topology:
    • Increased LCP-corr: UWS patients showed significantly higher LCP-corr values (p = 0.03) in the β1 band, indicating a greater tendency for local communities to form and interact.
  • Increased Clustering Coefficient: UWS patients also exhibited a higher clustering coefficient (p = 0.03), suggesting increased segregation within local network neighborhoods.
  • Increased Local Efficiency: UWS patients demonstrated higher local efficiency (p = 0.01), indicating enhanced fault tolerance within local network structures.
  • Small-Worldness: While both groups showed small-world properties, UWS patients had significantly lower SWw (p = 0.01) and SWw-E (p = 0.02) indices, suggesting a less efficient balance between segregation and integration.
  • Nodal Measures: While post-hoc analysis after multiple hypothesis testing correction did not yield significant results for single nodes, initial findings indicated increased nodal degree, betweenness centrality, and clustering coefficient in several frontal, parietal, and cingulate regions for UWS patients in the β1 band.
  • Discriminative Power: The study reported good performance for SWw (AUC = 0.81) and local efficiency (AUC = 0.80) in discriminating between UWS and MCS groups.
  • Clinical Correlation: Correlations between topological measures and CRS-R scores were generally poor and non-significant, suggesting that these network measures are more effective at group-level discrimination than at capturing fine-grained individual behavioral responsiveness.

Limitations and Future Work

  • Sample Size: The study acknowledges the relatively small sample size (n=25 analyzed datasets), which limits the generalizability of the findings.
  • EEG Limitations: The reliance on EEG, while advantageous for clinical accessibility, has inherent limitations in spatial resolution compared to fMRI. The study also notes that findings were primarily significant in the β1 frequency band, with other bands showing minor changes that did not reach statistical significance.
  • Thalamus Role: The study speculates on the role of the thalamus in sustaining network aberrations but could not directly investigate its specific involvement due to methodological constraints.
  • Nodal Level Significance: While nodal differences were observed, they did not reach statistical significance after multiple hypothesis testing correction, suggesting these findings require further investigation.

Future work could focus on larger, multi-center studies to confirm these findings, explore other frequency bands more thoroughly, and integrate EEG with other neuroimaging modalities (like fMRI) to gain a more comprehensive understanding of the brain's functional organization in DoC. Further research into the specific roles of the thalamus and the implications of altered local network organization in UWS for awareness and rehabilitation is also warranted.

Summary updated

Abstract

Consciousness arises from the functional interaction of multiple brain structures and their ability to integrate different complex patterns of internal communication. Although several studies demonstrated that the fronto-parietal and functional default mode networks play a key role in conscious processes, it is still not clear which topological network measures (that quantifies different features of whole-brain functional network organization) are altered in patients with disorders of consciousness. Herein, we investigate the functional connectivity of unresponsive wakefulness syndrome (UWS) and minimally conscious state (MCS) patients from a topological network perspective, by using resting-state EEG recording. Network-based statistical analysis reveals a subnetwork of decreased functional connectivity in UWS compared to in the MCS patients, mainly involving the interhemispheric fronto-parietal connectivity patterns. Network topological analysis reveals increased values of local-community-paradigm correlation, as well as higher clustering coefficient and local efficiency in UWS patients compared to in MCS patients. At the nodal level, the UWS patients showed altered functional topology in several limbic and temporo-parieto-occipital regions. Taken together, our results highlight (i) the involvement of the interhemispheric fronto-parietal functional connectivity in the pathophysiology of consciousness disorders and (ii) an aberrant connectome organization both at the network topology level and at the nodal level in UWS patients compared to in the MCS patients.
BibTeX
@article{cacciola2019functional,
  author = {Alberto Cacciola and Antonino Naro and Demetrio Milardi and Alessia Bramanti and Leonardo Malatacca and Maurizio Spitaleri and Antonino Leo and Alessandro Muscoloni and Carlo Vittorio Cannistraci and Placido Bramanti and Rocco Salvatore Calabrò and Giuseppe Anastasi},
  title = {{Functional Brain Network Topology Discriminates between Patients with Minimally Conscious State and Unresponsive Wakefulness Syndrome}},
  journal = {Journal of Clinical Medicine},
  year = {2019},
  doi = {10.3390/jcm8030306},
}
Figure