Electroencephalography functional connectivity-A biomarker for painful polyneuropathy

Area of Science:

• Neuroscience
• Medical Technology
• Pain Research

Background:

• Advanced electroencephalography (EEG) analysis is crucial for brain research.
• Diabetic polyneuropathy (DPN) pain assessment requires objective biomarkers.
• Resting-state EEG offers a data-driven approach for pain discrimination.

Purpose of the Study:

• To develop a predictive model using resting-state EEG to differentiate painful from non-painful DPN patients.
• To identify specific EEG functional connectivity patterns associated with DPN pain.
• To explore the correlation between EEG biomarkers and clinical pain parameters.

Main Methods:

• Acquired 3-minute resting-state EEG recordings from 180 DPN patients.
• Employed a combination of traditional, explanatory, and machine learning analyses.
• Identified top 10 functional bivariate connections per EEG band differentiating patient groups and correlated them with clinical pain data.

Main Results:

• Theta and beta EEG bands showed high accuracy (AUC 0.93 and 0.89) in discriminating painful DPN.
• Painful DPN patients exhibited significantly higher cortical functional connectivity in theta and alpha bands (p < 0.01).
• Individual functional connections positively correlated with reported pain levels across all frequency bands.

Conclusions:

• Resting-state EEG functional connectivity is a highly accurate biomarker for pain in DPN.
• This finding underscores the brain's significant role in the perception of clinical pain.
• The developed EEG analysis tool holds potential for application in other pain syndromes.